Prioritizing Infrastructure Investments in Panama: Pilot Application of the World Bank Infrastructure Prioritization Framework The World Bank and Ministry of Economy and Finance of Panama September 2015 Updated April 2016 Acknowledgements This report was authored by Darwin Marcelo, Cledan Mandri-Perrott, and Schuyler House on the World Bank Public-Private Partnerships Infrastructure team with valuable inputs and guidance from the Ministry of Economics and Finance of Panama, particularly Fabio Bedoya, Head of Investment Planning; Angela Dominguez, Coordinator of Investment Program Financing; Eliecer Lara, Project Evaluator; and Lourdes Arjona, Coordinator of Infrastructure. Important data inputs were provided by World Bank local consultants and infrastructure sectoral experts Ivan Estribi, Ivet Anguizola, Raul Bethancourt, Ricardo Cerrud, Wuadalquivir Fonseca, and Janice Campbell. 1 Introduction Infrastructure services are significant determinants of economic development, social welfare, trade, and public health. As such, they typically feature strongly in national development plans. While governments may receive many infrastructure project proposals, however, resources are often insufficient to finance the full set of proposals in the short term. Leading up to 2020, an estimated US$836 billion - 1 trillion will be required each year to meet growth targets worldwide (Ruiz-Nuñez & Wei, 2014; World Bank). Global estimates of infrastructure investments required to support economic growth and human development lie in the range of US$65-70 trillion by 2030 (OECD, 2006), while the estimated pool of available funds is limited to approximately US$45 trillion (B20, 2014). The past twenty years have also seen a shift towards decentralized infrastructure planning. Many subnational governments, regional entities, and sector agencies have been delegated responsibility for infrastructure planning promote local responsiveness, but responsibility for allocating funds often remains with a centralized finance agency (CFA). While constituencies may propose numerous projects, governments often have insufficient financial resources to implement the full suite of proposals. As such, government must make difficult decisions about which projects to select for implementation within a given investment period. This implies grappling with the relative efficiency and effectiveness of investments as well as project costs and benefits. The multiple considerations of project selection demand improved decision support frameworks that are sufficiently rigorous to accommodate multiple facets, yet practical to implement. Good practice suggests that economic and strategic project appraisals and feasibility studies provide a good basis for project prioritization via highest societal net present value as a ranking metric. The reality, however, is that capacity, resources, and time are often too short in supply to support extensive economic analysis across full project sets. Decision-makers may only have partial information on project costs and benefits, particularly since many are difficult to quantify and monetize. Thus, there is a need for evidence-based infrastructure decision support that is consistent and data-driven but pragmatic and responsive to the needs and current capacities of a government. The World Bank’s Infrastructure Prioritization Framework (IPF) responds directly to these demands by offering a systematic approach to infrastructure prioritization that places financial-economic and social-environmental factors at the forefront of decision-making. The framework is transparent and objective, following a clear step-wise approach, but allows space for deliberation in order to remain responsive to policy priorities. The IPF framework is differentiated in four ways from other approaches to infrastructure decision-making. First, it incorporates national policy goals, social and environmental sustainability considerations, and long-term development aims alongside financial and economic indicators. Second, it is predicated on parsimony and pragmatism. Third, it makes space for policy debate via criteria identification and the selection of projects from mid-priority categories. Fourth, it provides decision-makers with an intuitive, graphical interface upon which to compare alternative investment scenarios. IPF is not intended to replace traditional planning or policy analysis, which remain critical to identifying problems, assessing their relative importance or urgency, and selecting from amongst alternative solutions. These remain important pre-prioritization activities. IPF also 2 does not propose to substitute extensive economic analysis with more basic appraisal in the long term, or for major projects. Rather, IPF can be used employed as a catalyst to improve data collection in order to progress more sophisticated and extensive appraisal methods, including social cost-benefit analysis (SCBA). This report presents the IPF methodology and results of the pilot application to a select set of transport and water and sanitation projects in Panama. The report first gives background information on infrastructure prioritization in Panama, then follows with a description of the IPF in technical and implementation terms. Next, we present the results of the pilot and close with recommendations for implementing IPF to a wider set of proposed projects. Background In Panama, the current economic outlook and institutional supports spurred pilot application of the IPF. First, GDP growth and economic buoyancy in 2014 motivated an ambitious public investment program, accompanied by a high number of infrastructure project proposals to the Ministry of Economics and Finance. This was coupled with political commitment to narrow the deficit, and thus demanded that Government select some projects and postpone others for implementation during the following five-year strategic period. Institutionally, the application of a prioritization methodology was endorsed via the Government Strategic Plan 2015-2019, which called explicitly for systematic prioritization of public investments and consideration of socio-economic and environmental indicators. In keeping with the 2008 Social Fiscal Responsibility Law, the Government Strategic Plan sets forth action areas for investments in economic sectors, social, infrastructure, developments of people, environmental and governability over a five-year period. The 2015-2019 Plan is premised on achieving social equity, better standards of life for all Panamanians, and sustainable economic growth. The indicative plan is to be reviewed and updated annually to account for changes in forecasted income and expenses and the availability of capital to service project costs. As such, it recognizes that project execution is dependent on the availability of sufficient finance. The Strategic Plan accounts for public investments of approximately US$19.5 billion for the following five-year period. Of this planned total, 15.2% (US$2.96 billion) is marked for road projects, whereas 18.9% (US$3.69) is designated for water and sanitation. The Strategic Plan is also linked to the Five-Year Investment Plan, which specifies allocations to various Government ministries, agencies, and corporations for public expenditure. The 2014 World Bank Panama Country Partnership Framework (CPF) also calls for technical assistance to Panama on infrastructure prioritization (p. 19). In keeping with the Strategic Plan and CPF, in April 2015, the Ministry of Economy and Finance of Panama engaged the World Bank to undertake a technical advisory project to prioritize a select set of projects in the transport and water and sanitation sectors. Nineteen projects in transport and 35 in water and sanitation are the purview of the IPF pilot application herein. Developing Panama’s Infrastructure Infrastructure features heavily in Panama’s Government Strategic Plan 2015-2019, based on (a) an aim to reduce infrastructural constraints on per capita GDP (see Loayza et al, 2004, and Araujo, et al, 2014); and (b) recognition that national growth has stemmed largely from the 3 transport, commerce, and construction sectors. Construction has become a key economic driver due to investment (public and private) in residential and non-residential infrastructure (1.9% contribution over the past five years). The sectors also created most of the new employment from 2007 to 2012, particularly for low-skill workers (World Bank, 2015). While there is evidence of economic constraints from infrastructure, Panama has been ranked highly in international comparisons: it is one of the most competitive countries in the region on infrastructure developments, according to the Global Competitiveness Ranking prepared by the World Economic Forum, and the country has positioned itself as a key trade and logistics hub, naturally centered around the Panama Canal. Nevertheless, certain infrastructure subsectors have lagged, including urban connectivity and energy. With respect to the decision processes for infrastructure funding, a medium-term fiscal framework sets the budget ceiling for the central government, decentralized institutions, public companies and financial intermediaries. Institutions register programs and projects that require budgetary resources into a project bank (BP) maintained by the Ministry of Economy and Finance (MEF). The BP links pre-investment activities (e.g., proposal and appraisal) to budget execution. Assessments of proposed projects take into account the budget ceiling and the physical and financial performance of each institution. Based on this, the Directorate of Investment Programming (DPI) of the Ministry of Economy and Finance (MEF) provides a preliminary recommendation. This is then reviewed and modified by another MEF directorate, the Directorate of National Budget (DIPRENA), which assigns the final figure for each institution. Transport Roads are Panama’s weakest link with respect to trade infrastructure. Contrary to the country’s elevated position on international rankings for port and air transport infrastructure, Panama ranks 44th in the quality of road infrastructure (WEF, 2015), and road density is amongst the lowest in Central America. Rectifying this is particularly important, as Central America has yet to take full advantage of potential gains of regional integration and trade connectivity. The physical trade network limits bilateral trade, and logistics costs can be as much as 50% of the final price of traded goods. Estimates suggest that exports stand to double with integration (Marcelo, et al, 2010). Water In the water sector, three government bodies are responsible for policy and investment: (a) the Ministry of Health, via the Directorate of Water and Sanitation (DISAPAS), oversees sector planning and service delivery in towns and rural areas; (b) the National Water and Sewerage Institute / Instituto de Acueductos y Alcantarillados Nacionales (IDAAN) is responsible for urban areas and rural communities with over 1,500 inhabitants; and (c) the National Authority for Public Services (ASEP) supervises and regulates urban provision (World Bank, 2015). There are observed coordination problems, as infrastructure investments may be made in parallel, with no integrated sector investment plan. In this way, a structured prioritization process has much to offer in the way of organizing project selection, both within and across water agencies. One move towards better coordination, however, is in the development of a new Interagency Commission for Drinking Water and Sewerage (CIAPAS), involving the three agencies responsible for water and sanitation policies and investments. Regular meetings of working committees are held to address water and sanitation policy issues and disaster risks. 4 Whereas 94% of Panama’s population has access to an improved water source, only 73% have access to improved sanitation (World Bank b, 2015). Development needs in the sector are concentrated in the comarcas, 1 where many residents have little or no access to improved water supplies and sanitation. While public service quality, coverage, and reliability is generally lower in dispersed rural areas, services are particularly poor in the comarcas (World Bank, 2015). Poor water and sanitation coverage has major implications for child health, including the prevalence of diarrheal illness, and life expectancy overall. As such, if improving access to water and sanitation in the comarcas is identified as a key policy goal, this can be coded into future analysis via the inclusion in the social-environmental indicator. Figure 1. Access to running water in the dwelling, by concentration of indigenous peoples Source: World Bank, 2015 Poverty The concentration of poverty in rural indigenous areas has important implications for infrastructure development. 2 Indigenous residents of comarcas have markedly low access to basic services and infrastructure, and, whilst needs are also evident in urban areas and amongst other segments of the population, slow development and particularly low conditions in these areas cannot be overlooked. This requires that decision-makers deal with the inherently different economic performance of proposed projects in rural areas in some sectors in a way that does not penalize the communities where needs are greatest. Infrastructure Prioritization Framework The Infrastructure Prioritization Framework is a quantitative approach that synthetizes and displays financial and economic as well as social and environmental indicators at the project level and considers these alongside the public budget constraint for a sector. Results are displayed graphically to map comparative performance along these dimensions. While the IPF is quantitative in nature, it employs knowledge that is practice-based or political and opens space for deliberation in criteria and project selection. The approach recognizes 1. Comarca indígena or ‘comarcas’ in Panama are five indigenous regions, representing 20% of the national territory, which hold special administrative status alongside provinces. The comarcas are Ngobe-bugle and Campesino, Kuna Yalar, Embera-Wounan, Kuna de Madugandi, and Kuna de Wargandi. 2. 42% of Panama’s extremely poor residents live in the comarcas. 5 that project selection cannot be entirely dissociated from the political economy of a sector. Particular projects may be chiefly valued by governments and other stakeholders due to key policy goals which are non-economic in nature, or due to considerations that objective indicators cannot measure, such as upholding election promises, promoting social cohesion, or honoring culture. As such, the IPF accommodates policy and political responsiveness in two ways: through the selection of criteria (indicators) for assessment and by leaving a degree of freedom in decision-making through provision of two references for judgment (the indices). The IPF was created in response to observed government demand, including a needs for (a) improving infrastructure planning at national and sector levels; (b) considering large projects sets with scarce planning resources; (c) meaningfully addressing environmental and social factors; and (d) balancing analytical efficiency, derived from standardization, with policy and political responsiveness. Institutionalizing a systematic approach to prioritization is further justified by public demand for evidence, value, and legitimacy in infrastructure decision- making. These logics are detailed in Marcelo, et al, 2016. To be deemed legitimate and comprehensive, prioritization must be based on sound evidence that affords meaningful comparison. Many project comparisons are inherently technical; thus, IPF is designed to employ quantitative measures to the greatest extent possible to limit subjectivity. Furthermore, comprehensiveness requires that project comparisons make space for multiple policy goals and facets of project selection. This lends support to employing multi-criteria approaches, with criteria selected to reflect considerations of effectiveness and value, as well as sector and national policy goals. The IPF captures the strengths of multi-criteria decision approaches, but also allows use of inputs from cost-benefit analysis (CBA). 3 The process by which infrastructure is prioritized must also be administratively and politically feasible. This suggests that governments adopt the principle of parsimony – using the least amount of relevant information needed to inform a decision. Administrative feasibility means that approach can be implemented within limits of institutional capacity, cost, time, and data availability. And political feasibility accepts that prioritization cannot be totally unresponsive to political factors. In summary, most infrastructure policy contexts demand the reconciliation of highly technical and objective policy analysis with more political and practice-based inputs, all within the resource means of government. The IPF Process Implementing the IPF is relatively straightforward, following five steps: (1) selecting decision criteria; (2) gathering project indicator data; (3) calculating social-environmental and financial- economic indices; (4) plotting projects and budget limits; and (5) selecting projects (see Figure 2). In this section, we describe implementation in terms of these steps, with direct reference to the Panama pilot. An extensive technical description of the IPF methodology is also detailed in Marcelo, et al, 2016. 3. We recognize multiple approaches to investment decision-making, including social cost-benefit analysis (SCBA). For an extensive discussion of alternative approaches, see Marcelo, et al, 2016. Whereas SCBA requires fully monetized information on costs and benefits, IPF makes use of available inputs (e.g., financial CBA). IPF’s value-adds to CBA are in (a) directly treating non-marketed impacts in ‘natural’ units; (b) relieving the burden of making and justifying assumptions required to monetize benefits and costs; and (c) dealing directly with issues like equity and social justice. 6 Figure 2. IPF Process Map •Deliberation with decision I makers, experts, and key Select Criteria stakeholders Variables may be adjusted in the case of data problems or new information II •Source project data (CBA elements incorporated when available) Prepare Data •Transform / standardize data III Construct •Variables combined to create SEI and FEI Performance •PCA applied when objectivity is preferred Indices IV •Plot SEI and FEI coordinates Plot Visual •Quadrants defined by budget constraint Interface V •Based on informed political and Select Projects technical debate •Quadrant A = high priority Source: Marcelo, et al, 2016 One important pre-analytical step is the delineation of proposed projects to which the IPF will apply. This may include the full universe of projects proposed to Government, or a select set. In the case of the Panama pilot, experts from the Ministry of Economy and Finance (MEF), in consultation with participants from the Ministry of Public Works / Ministerio de Obras Públicas (MOP), Ministry of Health / Ministerio de Salud (MINSA), and the National Water and Sewerage Institute / Instituto de Acueductos y Alcantarillados Nacionales (IDAAN), a consultant in charge of the Government’s Strategic Plan, and experts from the World Bank, randomly selected 34 and 19 projects in the water and urban mobility sectors, respectively. Basic information for these projects was obtained from the Banco de Proyectos, the platform of the National Public Investment System (SINIP), where all public investments carried out by state agencies are recorded at the national level. Only projects above US$ 5 million were included in this exercise, as they are bound to meet the data requirements of Article 12 of Law 25 (amending Article 23 of Law 34 on Social and Fiscal Responsibility). This law indicates that investment projects with proposed costs equal to or greater than 0.1% of the General State Budget must include studies on costs and social benefits, whereas projects less than 0.1% but more than five million dollars (US$ 5,000,000) require at least prefeasibility studies. The 19 transport projects included in the analysis are listed in Annex 1, along with project costs and the raw data on indicator values. The 35 water and sanitation projects included in the analysis are listed in Annex 2, along with project costs and the raw data on indicator values. Participants from the Institute of Aqueducts and Sewers (IDAAN), Ministry of Public Works (MOP), and Ministry of Health (MOH) were included in the implementation stages. Step 1. Identifying criteria The first step in applying the IPF is to select the criteria used to compare projects. The selection of variables seeks to preserve the principle of parsimony. Accordingly, this methodology 7 requires a minimum level of relevant information to assess the various expected outcomes of proposed infrastructure projects. The selection of variables may differ amongst application contexts, based on government policy goals (e.g., particular sectorial, social, and environmental aims) and stakeholder consultations, but will generally include indicators of value, efficiency, and social and environmental impact. This step is an opportunity to leverage professional knowledge and allow policy-makers, experts, and other key stakeholders to reach consensus on the decision factors most important to project selection. In this way, the step crystallizes the Government’s agreed goals of infrastructure development. In the Panama pilot, the initial list of variables was agreed during a technical workshop including participants from the World Bank team, Ministry of Economy and Finance (MEF) technical staff, National Water Supply and Sanitation Administration / Institute of Aqueducts and Sewers (IDAAN), Ministry of Health (MOH), Ministry of Public Works (MOP), and Ministry of Government (MINGOB). The variables are organized into two categories: social-environmental and financial-economic. The indicators originally selected were as follows: Social-Environmental Indicators Financial-Economic Indicators • Beneficiaries (people served) • Internal rate of return • Jobs created • Multiplier effects • Negatively affected (people and costs) • Externalities (e.g., avoided diseases) • Poverty (coverage map) • Implementation risks • Environmental footprint (positive, neutral, negative) Infrastructure projects are meant to improve quality of life. A number of direct social and environmental benefits are relevant, including improved access to public services and job and income opportunities created during construction and operation. These benefits come at a cost, however. Engineering works may require clearing forested areas, polluting and endangering natural environments, or resettling communities. The IPF directly considers relevant social and environmental benefits and costs via indicators used to construct a social and environmental index (SEI). In Panama, the SEI initially consisted of five indicators: the number of direct beneficiaries; direct jobs created; people affected by repurposing of land use; poverty; and the environmental footprint (categorized as negative, neutral, or positive). The financial and economic viability and fiscal burden of a project is a central issue in all infrastructure decision-making. Four indicators were initially selected to comprise the financial-economic index (FEI): the internal rate of return, economic multiplier effects, monetizable externalities, and implementation risk. Adjusting criteria variable specification One key lesson drawn from the Panama pilot was that a need to adjust some of the original indicators may arise. An indicator may be found to be analytically problematic due to lack of sufficient data, calculation problems, or other issues such as imprecision in the variable specification or specifications that do not capture policy goals. This is an iterative process, and indicator problems are likely to be discovered during data collection (Step 2) or index calculation (Step 3). In the case of Panama, the major constraints that emerged were (1) 8 variable incalculability, (2) unavailability of data, (3) insufficient precision in variable specification, and (4) sector-specific needs. These are described in the following table. Table 1. Indicator variable adjustments Indicator change Reason for adjustment 1. ‘Number affected’ eliminated from SEI No data available 2. ‘Environmental footprint’ eliminated from SEI Indicator found too imprecise (1-3) to reflect scale of differences between projects. 3. Transport SEI variables controlled for project Beneficiaries and jobs created per dollar gives an indication of size (cost); original SEI variables transformed relative efficiency of projects. In contrast, SEI variables for water and expressed as ratios of cost (e.g., number projects were not adjusted, as doing so would have privileged urban of beneficiaries / jobs created per dollar) projects over rural. 4. Absolute number of potential poor receiving Poverty rates are higher in less densely populated areas, but in service used instead of poverty ratio (%) to absolute terms, poor are mainly concentrated in urban areas. The calculate SEI use of poverty rates could favor projects with lower impacts to poor when using PCA. Using absolute values corrected the problem. 5. FEI for both sectors limited to benefit-cost Insufficient data for multiplier effects, externalities, and ratio implementation risk IRR not mathematically calculable in many cases due to very low monetized benefits An important adjustment was made to the calculation of FEI. The FEI was originally expected to include the internal rate of return (IRR) and/or economic rate of return (ERR) of projects, depending on data availability, in addition to several other indicators. Since many of these investments were proposed for projects with no direct monetary benefits and largely indirect economic effects (i.e. mainly projects with a large public good component), however, the calculation of IRR would have produced unrealistic or incalculable results. The alternative would have been to account for all indirect positive effects and estimate benefits. This would require data on monetized benefits for such effects, which was not available. For these reasons, we relied on the NPV (assuming given discount rates) of available information to calculate benefit-cost ratios (BCRs). The BCRs also allowed us to control for project size and avoid penalizing projects with higher costs but possibly higher benefits. Step 2. Gathering and transforming data The second step is to gather and Table 2. Summary of consultant time for Panama pilot transform data to calculate social- Consultant Sector Days environmental and financial-economic 1 Urban Transport 33 scores for each project. A team of six local 2 Urban Transport 20 consultants with sectorial expertise 3 Operations/Transport 7 gathered data under the supervision of 4 Water/Sanitation 40 World Bank staff (Table 2). Consultants 5 Water/Sanitation 20 6 Water/Sanitation 15 gathered information from existing Total 135 project information sheets and made Average days per consultant 22.5 additional calculations based on research. Data was entered into an Excel environment, which recorded raw project data associated with each variable, for each project under study. 4 A simple Excel environment could be constructed in the future to assist the population of a prioritization database. An example data entry form 4. For analytical purposes, we also coded some projects as ‘outliers’, since their extreme values would have distorted the calculation of weight via PCA. 9 is included in Annex 3, and an example populated data table for Water and Sanitation Projects is shown in Table 3 below. Table 3. Water and Sanitation Project Excel Data Table PCA Weighting Equal Weighting ID Cost BEN EMP POOR BEN EMP POOR SIE SEI SIE SEI STD STD STD STD STD Project 1 17,650,597 206481 125 74333 0.47 -0.91 1.10 0.87 36.76 0.38 40.33 Project 2 874935 138 166238 3.76 -0.65 3.35 3.73 84.08 Project 3 5,887,649 315019 156 63004 1.01 -0.29 0.82 1.19 41.43 0.89 46.97 Project 4 5,991,593 67265 156 36996 -0.21 -0.29 0.19 -0.09 22.93 -0.18 32.97 Project 5 7,450,110 36512 98 13144 -0.36 -1.46 -0.40 -0.86 11.66 -1.28 18.64 Project 6 5,971,133 43939 98 8348 -0.33 -1.46 -0.52 -0.92 10.85 -1.33 18.03 Project 7 30,000,000 53247 208 13844 -0.28 0.75 -0.38 -0.28 20.12 0.05 36.06 Project 8 6,800,000 3723 191 1378 -0.53 0.41 -0.69 -0.74 13.49 -0.46 29.35 Project 9 5,250,000 6813 189 1363 -0.51 0.37 -0.69 -0.74 13.50 -0.48 29.16 Project 10 5,000,000 206481 153 74333 0.47 -0.35 1.10 1.00 38.67 0.70 44.58 Project 11 15,300,000 16636 152 4658 -0.46 -0.37 -0.61 -0.82 12.28 -0.83 24.52 Project 12 60,000,000 116000 148 25520 0.03 -0.45 -0.09 -0.15 21.97 -0.30 31.45 Project 13 9,000,000 47401 103 25123 -0.31 -1.36 -0.10 -0.60 15.45 -1.02 22.01 Project 14 5,000,000 3905 102 1406 -0.53 -1.38 -0.68 -1.15 7.46 -1.49 15.87 Project 15 50,000,000 156698 239 86184 0.23 1.38 1.39 1.43 44.96 1.73 57.95 Project 16 9,980,000 161470 103 41982 0.25 -1.36 0.31 0.07 25.16 -0.46 29.36 Project 17 40,000,000 31223 238 21544 -0.39 1.36 -0.19 -0.08 22.96 0.45 41.21 Project 18 5,000,000 9788 186 5579 -0.50 0.31 -0.58 -0.67 14.47 -0.44 29.59 Project 19 10,000,000 15339 208 4142 -0.47 0.75 -0.62 -0.57 15.89 -0.19 32.87 Project 20 8,000,000 7006 203 1962 -0.51 0.65 -0.67 -0.66 14.61 -0.30 31.40 Project 21 7,000,000 2890 203 1329 -0.53 0.65 -0.69 -0.68 14.25 -0.32 31.13 Project 22 20,000,000 20420 234 11231 -0.44 1.28 -0.44 -0.31 19.64 0.22 38.30 Project 23 50,000,000 220779 243 50779 0.54 1.46 0.52 1.07 39.73 1.46 54.41 Project 24 50,000,000 161470 243 41982 0.25 1.46 0.31 0.72 34.67 1.16 50.58 Project 25 5,000,000 5181 186 2901 -0.52 0.31 -0.65 -0.73 13.59 -0.49 28.92 Project 26 18,000,000 7980 184 3032 -0.51 0.27 -0.65 -0.73 13.63 -0.51 28.75 Project 27 60,000,000 64856 234 17511 -0.23 1.28 -0.29 -0.06 23.35 0.44 41.11 Project 28 42,825,811 50694 210 10139 -0.29 0.80 -0.47 -0.34 19.22 0.02 35.59 Project 29 24,397,000 47401 208 25123 -0.31 0.75 -0.10 -0.11 22.59 0.20 37.93 Project 30 5,413,130 2313 118 1411 -0.53 -1.05 -0.68 -1.08 8.47 -1.31 18.23 Project 31 12,674,150 31537 162 11669 -0.39 -0.17 -0.43 -0.61 15.40 -0.57 27.88 Project 32 9,000,000 7364 125 2651 -0.51 -0.91 -0.65 -1.01 9.50 -1.20 19.71 Project 33 5,500,000 7602 98 2129 -0.51 -1.46 -0.67 -1.15 7.55 -1.52 15.53 Project 34 5,000,000 31223 103 21544 -0.39 -1.36 -0.19 -0.72 13.79 -1.12 20.75 Project 35 5,000,000 64856 148 17511 -0.23 -0.45 -0.29 -0.46 17.51 -0.56 28.08 The Social and Environmental Index (SEI) and the Financial and Economic Index (FEI) are each built on sets of quantitative variables combined into one index via an additive model. To condense disparate data types and scales of measurement into indices, three data transformations are required. One must (a) transform ordinal qualitative data into usable quantitative scalar data; (b) standardize criteria to isolate the various units of measure; and (c) establish weights for each criterion in the additive model. 10 Because all of the variables selected were already quantitative, we did not have to perform the first transformation. We did, however, standardize the numerical values via a typical standardization formula. The standard score of a raw score is − = , where is the sample mean and is the standard deviation of the variable j. The standardization formulas are coded into the Excel platform. Step 3. Constructing the SEI and FEI Indices are used to combine information from multiple variables into one composite indicator. In IPF, there are two variable classes that comprise the social-environmental index (SEI) and the financial-economic index (FEI). In the case of Panama, since the FEI was comprised of a single component (the standardized benefit-cost ratio), its calculation was based on the standardization and normalization steps (see Step 2) only. The SEI, on the other hand, was constructed by an additive model that combined variables with associated weights. The function may be expressed as = 1 × () + 2 × () + 3 × (), where weights, , are associated with each of the social-environmental indicators. Principal Component Analysis (PCA) was used to determine the weights of each variable in the additive function. PCA is an information reduction procedure that seeks redundancies in a set of variables. These redundancies can be expressed as linear combinations or ‘principal components’ of the variables comprising the set, where each principal component is a weighted average of the original indicators. 5 The first principal component corresponds to the linear combination of variables that retains the maximum information of the original data set (Pearson, 1901). One of the main characteristics of PCA is the ability to calculate coefficients based solely on the statistical relationship between variables. This is particularly useful when there is a preference to objectively assign weights. Nevertheless, other methods such as equal weighting or expert-decided weighting may also be employed for reference. 6 After the SIEs and FIEs are calculated for each project, they are normalized and rescaled to generate scores between 0 and 100 for each project’s index score. The rescaled score is − ( ) = × 100 − where is the minimum value for variable Z and is the maximum value. These rescaled scores are used as the SEI and FEI scores for plotting in Step 4. 5. The coefficients, or weights, associated with variables in each principal component are those that maximize the variance of each. The notation for the first principal component is: = 1 2 +2 2 +…+ , where denotes each observation and denotes the weight for the nth variable . The coefficients of each ‘first principal component’ are taken as the weights associated with each variable . 6. The simplest mode of weighting is equality, where all criteria are equally considered. An alternative method is negotiated expert guidance, where a panel of decision-makers decides weights based on experience. 11 Transport Index Scores The SEI index scores were based on PCA-determined weights of .650, .439, and .633 associated with beneficiaries per dollar (BEN_COS), employment per dollar (EMP_COS), and poor people served per dollar (POOR_COS), respectively. The results are shown in Figure 3. Figure 3. Transport Social-Environmental Index, PCA weighting 100 90 80 70 60 50 40 30 20 10 0 Project 12 Project 15 Project 14 Project 16 Project 18 Project 11 Project 13 Project 10 Project 19 Project 17 Project 1 Project 4 Project 5 Project 9 Project 8 Project 7 Project 2 Project 6 Project 3 Figure 4 shows results of the transport sector FEI calculations. Note that the FEI reflects the relative importance of projects in terms of the benefit-cost ratio. Figure 4. Transport Financial-Economic Index 100 90 80 70 60 50 40 30 20 10 0 Project 15 Project 10 Project 14 Project 12 Project 3 Project 11 Project 16 Project 9 Project 13 Project 19 Project 2 Project 8 Project 7 Project 1 Project 18 Project 6 Project 5 Project 17 Finally, a sensitivity analysis was performed on the SEI to determine the effect of using other Project 4 weighting schemes besides PCA on the calculated SEI scores for each project. This is described in Step 4. The alternative tested for Transport was equal weighting of SEI variables. Figure 5 shows that project rankings by SEI changes very minimally. Figure 5. Transport Social-Environmental Index, Equal Weighting 100 90 80 70 60 50 40 30 20 10 0 Project 12 Project 15 Project 14 Project 18 Project 16 Project 11 Project 13 Project 10 Project 19 Project 17 Project 1 Project 4 Project 5 Project 9 Project 7 Project 8 Project 2 Project 6 Project 3 12 0 10 20 30 40 50 60 0 0 10 20 30 40 50 60 70 80 90 10 20 30 40 50 60 70 80 90 100 100 Project 15 Project 3 Project 15 Project 3 Project 10 Project 23 Project 23 Project 35 Project 24 Project 10 Project 6 Project 3 Project 1 Project 1 Project 10 Project 24 Project 16 Project 17 Project 16 Project 5 Project 27 Project 27 Project 13 Project 1 Project 17 Project 15 Project 22 Project 4 Project 12 Project 29 Project 29 Project 4 Project 7 Project 12 Project 7 Project 28 Water and Sanitation Index Scores Project 7 Project 31 Project 4 Project 22 Project 34 Project 19 Project 28 Project 23 Project 12 Project 35 Project 29 Project 20 Project 19 Figure 6. Water and Sanitation Financial-Economic Index Project 28 Project 21 Project 13 Project 24 Project 18 Project 31 Project 33 Project 16 Project 27 Project 8 Project 20 Project 19 Project 9 Project 18 Project 14 Project 25 Project 21 Figure 4. Water and Sanitation Social-Environmental Index, PCA weighting Project 11 Project 26 Project 34 Figure 5. Water and Sanitation Social-Environmental Index, Equal Weighting Project 18 Project 35 Project 26 Project 32 Project 31 Project 25 scoring projects remained high, and low scoring ones remained low. Project 17 Project 11 Project 9 Project 9 Project 13 Project 8 Project 25 Project 34 Project 11 Project 22 Project 32 Project 5 Project 20 Project 5 Project 6 respectively, for beneficiaries (BEN), jobs created (EMP), and poor served (POOR). Project 8 Project 30 Project 32 Project 26 Project 6 Project 30 Project 30 Project 14 Project 33 Project 21 Project 33 Project 14 (Figure 7). In this case, the ordering of projects changed more than with transport, but high 13 As with transport, an alternative SEI calculation based on equal was used for sensitivity analysis Similarly, SEI scores for water and sanitation projects used PCA weights of .687, .233, and .688, Determining the Budget Limit The SEI and FEI rankings are used to determine the set of projects fundable under the budget limits for the visual interface to follow in Step 4. Budget limits are mapped separately for SEI and FEI rankings, but using the same budget limit value. In other words, the limits are imposed on each index ranking, according to the hypotheticals that projects would be entirely selected according to SEI or to FEI, respectively. The process is simple: the available budget is ‘allocated’ to the top-ranking project on the SEI, followed by the second, third, and so on, until the available resources are exhausted. For example, the total cost of proposed water and sanitation projects (excluding Project 2) was US$622,091,173. Based on recommended figures from the MEF (DIPRENA and DPI), in turn based on the Draft Annual Budget, the limit for water and sanitation was assumed to be 55% of the total cost of proposed projects, equaling $342,150,145. Considering the costs of the top- ranking SEI projects 15, 3, 23, and so on, the budget would be exhausted after Project 29, at which the cumulative expenditure would be $318,906,839. Since Project 12 costs $60,000,000, there is insufficient funding to include it. Figure 7. SEI-ordered water and sanitation projects within the budget limit 60 budget limit 50 40 30 20 10 0 Project 15 Project 23 Project 10 Project 24 Project 16 Project 27 Project 17 Project 29 Project 12 Project 22 Project 28 Project 35 Project 19 Project 13 Project 31 Project 20 Project 18 Project 21 Project 34 Project 26 Project 25 Project 11 Project 32 Project 30 Project 33 Project 14 Project 3 Project 1 Project 4 Project 7 Project 9 Project 8 Project 5 Project 6 The same is done for the FEI. For water and sanitation, the last project that could be funded was also Project 29, with a cumulative expenditure of $304,002,232. The cumulative totals are different because the ranking of projects according to each index are different. Figure 8. FEI-ordered water and sanitation projects within the budget limit 100 90 budget limit 80 70 60 50 40 30 20 10 0 Project 10 Project 35 Project 16 Project 13 Project 15 Project 12 Project 31 Project 34 Project 23 Project 29 Project 28 Project 24 Project 33 Project 27 Project 19 Project 14 Project 11 Project 18 Project 32 Project 17 Project 25 Project 22 Project 20 Project 26 Project 30 Project 21 Project 3 Project 6 Project 1 Project 5 Project 4 Project 7 Project 9 Project 8 For transport, the budget constraint was estimated based on historical allocation of about 60% of requested funding. This resulted in an estimated $139,586,000 budget limit. 14 Step 4. Using the Visual Interface: Plotting Results on the Investment Frontier A ‘good’ project, in terms of financial and economic performance, may nevertheless be undesirable from a social and environmental perspective, and vice versa. As such, decision- makers must consider projects along both dimensions. Projects can be compared by their respective SEI and FEI scores on a visual interface called the Infrastructure Prioritization Matrix. Step 4 involves plotting each project on a Cartesian plane, with axes defined by the SEI and FEI. In Figure 9, each point represents a proposed transport project, whose location on the plane is determined by (x,y) coordinates defined by the (FEI, SEI) score pair. Once projects are plotted, the budget limit described above is imposed onto the plane, perpendicular to each axis, at the point where the budget would exhausted if project funding was determined solely according to each index. Since this is done along each axis, four quadrants are formed. In Figures 9 and 10, the red dotted lines represent the budget constraint for the sector that delineates the four quadrants. Projects that fall inside the budget constraint along each axis (i.e., above the dotted line, respective to the y or SEI axis, and to the right of the dotted line, respective to the x or FEI axis) represent the Investment Possibilities Set for each dimension. The set of projects in Quadrant A are those that fall in the Investment Possibilities Set for both SEI and FEI. These are categorized as ‘High Priority’ projects. Projects in Quadrant B are categorized as ‘High Priority SEI’ projects, and Quadrant C projects are categorized as ‘High Priority FEI’ projects. For transport, an estimated budget of $139,585,800 is available. If the high priority projects are implemented first, totaling US$77,430,000 (see Table 4), a remainder of US$62,155,800 is available for other proposed projects that may be selected from quadrants B or C depending on policy priorities. In the case of the water sector, the budget available for water and sanitation projects is US$342,150,145. Table 5 shows that, after funding high priority projects for a total of US$144,509,839, a remaining $197,640,306 is available for other prioritization categories. 15 Figure 9. Transport Infrastructure Prioritization Matrix 100 P12 90 P15 80 B P14 A 70 P1 P16 P11 60 P18 P4 50 SEI P13 40 P19 P10 30 P17 20 P8 P9 P7 P5 10 P2 0 P6 D P3 C 0 10 20 30 40 50 60 70 80 90 100 FEI Table 4. IPM Transport Project Categorization High priority projects Cost Project 2 Rehabilitación Camino Tortí-Chimán-Gonzalo Vásquez 45,000,000 Project 9 Rehabilitación Calles de Penonomé (Vista Hermosa) 4,880,000 Project 10 Rehabilitación Calles de Playa Leona (La Mitra hacia Peñas Blancas y Paso Arenas) 4,720,000 Project 11 Rehabilitación Calles de La Chorrera 1,600,000 Project 12 Rehabilitación Calles de Chilibre 4,500,000 Project 13 Rehabilitación Calles del Distrito de Aguadulce 4,200,000 Project 14 Rehabilitación Calles de Colón (Casco Antiguo y Arco Iris) 4,000,000 Project 15 Rehabilitación Calles del Distrito de Bugaba 2,000,000 Project 16 Rehabilitación Calles de Santiago (2015) 4,050,000 Project 19 Rehabilitación Carretera Lagarterita-Los Hules-Las Pavas -Lagartera Grande y Mejora del 2,480,000 Puente TOTAL 77,430,000 High priority social-environmental projects Project 1 Rehabilitación Carretera de Anachucuna 1,800,000 Project 4 Rehabilitación carretera de Irgandi (para llegar a la Comunidad de Playón Chico) 2,997,000 Project 5 Rehabilitación Carretera CPA-Calabacito 4,800,000 Project 7 Rehabilitación Carretera Tambo-Las Marías y Construcción de Puentes Vehiculares (5) 8,948,000 Project 8 Rehabilitación Carretera Loma Del Naranjo-Tres Quebradas 5,400,000 Project 17 Construcción Camino Quebrada Honda 8,000,000 Project 18 Rehabilitación Carretera Juay-Quebrada del Loro -Cañazas 1,600,000 TOTAL 33,545,000 High priority financial-economic projects Project 3 Rehabilitación Carretera Las Tablas-Los Asientos-Cañas-Cacao 58,668,000 TOTAL 58,668,000 Lower priority projects Project 6 Rehabilitación Carretera Circunvalación Garachiné-La Palma 63,000,000 TOTAL 63,000,000 16 Figure 10.Water and Sanitation Infrastructure Prioritization Matrix 50 45 P15 B A P3 40 P23 P10 P1 P24 35 30 25 P27 P29 P4 P16 P17 SEI P12 P22 P28 20 P7 P35 P19 15 P13 P21 P18 P26 P25 P31 P5 P11 P34 10 P32 P6 C P20 P30 P33 P14 5 D 0 0 10 20 30 40 50 60 70 80 90 100 FEI Table 5. IPM Water and Sanitation Project Categorization High priority projects Cost Project 1 Mejoramiento sistemas de agua potable y saneamiento- distrito de Colón 17,650,597 Project 3 Construcción de línea de conducción del Tanque Ameglio - Red de distribución del Sector 3 5,887,649 Project 4 Mejoramiento obras complementarias red de distribución - Área Metropolitana (Línea de 5,991,593 Conducción Pacora - Tanara - Tataré Project 10 Construcción de línea paralela Sabanitas, Planta - Cuatro Altos 5,000,000 Project 15 Estudio, diseño y construcción para una nueva planta potabilizadora del Río Bayano - 50,000,000 Project 16 Construcción de obras complementarias para red de distribución de La Chorrera (Estación de 9,980,000 bombeo y proyectos complementarios La Chorrera) Project 23 Diseño y construcción del sistema de alcantarillado sanitario de Arraiján 50,000,000 Project 29 Ampliación y mejoras al sistema de alcantarillado sanitario de Puerto Armuelles - CAF - II 24,397,000 FASE TOTAL 144,509,839 High priority social-environmental projects Project 17 Construcción Sistema de alcantarillado de Changuinola - 40,000,000 Project 24 Mejora Integral del sistema de alcantarillado sanitario de 50,000,000 Project 27 Santiago Construcción del sistema de alcantarillado sanitario (Sistema de recolección y 60,000,000 tratamiento de aguas residuales) CAF - II FASE TOTAL 150,000,000 17 High priority financial-economic projects Project 5 Construcción de mejoras a la red de distribución de Chorrillo y Santa Ana 7,450,110 Project 6 Mejoramiento al Acueducto de San Francisco 5,971,133 Project 7 Construcción de línea de conducción de La Chorrera - Capira, (Sectores 7 y 8) 30,000,000 Project 12 Mejoramiento integral al abastecimiento de agua de David 60,000,000 Project 13 Mejoramiento al sistema de agua potable de Puerto Armuelles 9,000,000 Project 29 Ampliación y mejoras al sistema de alcantarillado sanitario de Puerto Armuelles - CAF - II 24,397,000 FASE Project 31 , Construcción del sistema de abastecimiento de agua potable de Chorro Blanco, Alanje - 12,674,150 Boquerón II Etapa Project 34 Nueva toma de agua cruda en Bonyic, Changuinola 5,000,000 Project 35 Ampliación de la planta potabilizadora - Santiago 5,000,000 TOTAL 159,492,393 Lower priority projects Project 8 Construcción del sistema de alcantarillado sanitario de Parita 6,800,000 Project 9 Construcción del Sistema de Alcantarillado Sanitario de Chilibre Centro 5,250,000 Project 11 Construcción de nueva planta potabilizadora de 10 MGD de Dolega y sectores aledaños 15,300,000 Project 14 Mejoras al sistema de abastecimiento y distribución de agua potable de Santa Marta, Santo 5,000,000 Domingo Project 17 Construcción Sistema de alcantarillado de Changuinola 40,000,000 Project 18 Construcción de mejoras al Sistema de Alcantarillado Sanitario de Antón - 5,000,000 Project 19 Construcción del sistema de alcantarillado sanitario de Boquete 10,000,000 Project 20 Construcción del sistema de alcantarillado sanitario de Ocú - 8,000,000 Project 21 Construcción de alcantarillado sanitario de Macaracas 7,000,000 Project 22 Diseño y Construcción Sistema de alcantarillado sanitario de Chepo 20,000,000 Project 25 Diseño y construcción del sistema de alcantarillado sanitario de Capira 5,000,000 Project 26 Construcción del sistema de alcantarillado sanitario de Bejuco - Chame 18,000,000 Project 28 Ampliación y mejoras del sistema de alcantarillado sanitario de Chitré y alrededores - 42,825,811 (Ejecuta el PAN) Project 30 Construcción del sistema de alcantarillado sanitario de Metetí. 5,413,130 Project 32 Captación y ampliación de la planta potabilizadora de Isla Colón - CAF - II FASE 9,000,000 Project 33 Mejoramiento al sistema de agua potable de El Valle de Antón 5,500,000 TOTAL 208,088,941 18 Considering Subsectors: Separating Water and Sanitation A look at the project rankings and Figure 10 shows that water projects tend to score higher on FEI than sanitation projects. As such, the ranking of projects together results in a matrix that classifies no sanitation projects as ‘High Priority’ nor ‘High Financial-Economic Priority’. The three ‘High Social-Environmental Priority’ projects, conversely, are all sanitation projects. This observation reaffirms the need to target investments specifically to sanitation to ensure that much-needed resources are secured for implementation of sanitation improvements. By allocating resources by subsector, the government can identify the sanitation projects that are positioned in the high priority space and carve out budget specifically for the subsector. To determine additive weights for SEI, we again performed Principal Components Analysis, but separately for the water and sanitation project sets. This would be done for FEI also, if more than one input variable was involved. Table 6. Revised Water and Sanitation Project SEI PCA Weights BEN EMP POOR Combined water and sanitation PCA weighting 0.687 0.233 0.688 Water only PCA weighting 0.625 0.42 0.657 Sanitation only PCA weighting 0.598 0.516 0.613 Next, we estimated separate subsector budget constraints based on 55% allocation of total subsector project costs for each. And finally, we created separate Infrastructure Prioritization Matrices that identified two sets of High Priority projects based on the separate analysis. Whereas no sanitation projects were classified as High Priority in combined analysis, separating the subsectors promotes sanitation Projects 23 and 24 to High Priority status. The analysis does not change the ranking of projects within each subsector. Figure 10. Water Projects: Infrastructure Prioritization Matrix 60 P15 50 P3 P10 40 P1 P7 P4 30 P12 P16 SEI P35 P11 P31 P13 20 P34 P5 P32 P6 P33 P14 10 0 0 10 20 30 40 50 60 70 80 90 100 FEI 19 Table 7. IPM Water Project Categorization High priority projects Cost Project 1 Mejoramiento sistemas de agua potable y saneamiento- distrito de Colón 17,650,597 Project 3 Construcción de línea de conducción del Tanque Ameglio - Red de distribución del Sector 3 5,887,649 Project 10 Construcción de línea paralela Sabanitas, Planta - Cuatro Altos 5,000,000 Project 15 Estudio, diseño y construcción para una nueva planta potabilizadora del Río Bayano - 50,000,000 TOTAL 78,538,246 High priority social-environmental projects Project 4 Mejoramiento obras complementarias red de distribución - Área Metropolitana (Línea de 5,991,593 Conducción Pacora - Tanara - Tataré Project 7 Construcción de línea de conducción de La Chorrera - Capira, (Sectores 7 y 8) 30,000,000 TOTAL 35,991,593 High priority financial-economic projects Project 5 Construcción de mejoras a la red de distribución de Chorrillo y Santa Ana 7,450,110 Project 6 Mejoramiento al Acueducto de San Francisco 5,971,133 Project 13 Mejoramiento al sistema de agua potable de Puerto Armuelles 9,000,000 Project 16 Construcción de obras complementarias para red de distribución de La Chorrera 9,980,000 Project 35 Ampliación de la planta potabilizadora de Santiago 5,000,000 TOTAL 37,401,243 Lower priority projects Project 11 Construcción de nueva planta potabilizadora de 10 MGD de Dolega y sectores aledaños 15,300,000 Project 12 Mejoramiento integral al abastecimiento de agua de David 60,000,000 Project 14 Mejoras al sistema de abastecimiento y distribución de agua potable de Santa Marta, Santo 5,000,000 Domingo Project 31 Construcción del sistema de abastecimiento de agua potable de Chorro Blanco, Alanje – 12,674,150 Boquerón - II Etapa Project 32 Captación y ampliación de la planta potabilizadora de Isla Colón - CAF - II FASE 9,000,000 Project 33 Mejoramiento al sistema de agua potable de El Valle de Antón 5,500,000 Project 34 Nueva toma de agua cruda en Bonyic Changuinola 5,000,000 TOTAL 253,268,229 20 Figure 11. Sanitation Projects: Infrastructure Prioritization Matrix 100 P23 90 P24 80 70 P17 P27 60 P29 P22 P28 50 P21 P20 SEI P19 40 P8 P9 P18 P26 P25 30 P30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 FEI Table 8. IPM Sanitation Project Categorization High priority projects Cost Project 23 Diseño y construcción del sistema de alcantarillado sanitario de Arraiján 50,000,000 Project 24 Mejora Integral del sistema de alcantarillado sanitario de La Chorrera 50,000,000 TOTAL 100,000,000 High priority social-environmental project Project 27 Construcción del sistema de alcantarillado sanitario de Santiago (Sistema de 60,000,000 recolección y tratamiento de aguas residuales) CAF - II FASE TOTAL 60,000,000 High priority financial-economic projects Project 28 Ampliación y mejoras del sistema de alcantarillado sanitario de Chitré y alrededores - 42,825,811 (Ejecuta el PAN) Project 29 Ampliación y mejoras al sistema de alcantarillado sanitario de Puerto Armuelles - CAF-II 24,397,000 FASE TOTAL 42,825,811 Lower priority projects Project 8 Construcción del sistema de alcantarillado sanitario de Parita 6,800,000 Project 9 Construcción del Sistema de Alcantarillado Sanitario de Chilibre Centro 5,250,000 Project 17 Construcción Sistema de alcantarillado de Changuinola 40,000,000 Project 18 Construcción de mejoras al Sistema de Alcantarillado Sanitario de Antón 5,000,000 Project 19 Construcción del sistema de alcantarillado sanitario de Boquete 10,000,000 Project 20 Construcción del sistema de alcantarillado sanitario de Ocú 8,000,000 Project 21 Construcción de alcantarillado sanitario de Macaracas 7,000,000 Project 22 Diseño y Construcción Sistema de alcantarillado sanitario de Chepo 20,000,000 Project 25 Diseño y construcción del sistema de alcantarillado sanitario de Capira 5,000,000 Project 26 Construcción del sistema de alcantarillado sanitario de Bejuco - Chame 18,000,000 Project 30 Construcción del sistema de alcantarillado sanitario de Metetí - 5,413,130 TOTAL 130,463,130 21 Sensitivity Analysis To improve the robustness of prioritization results, a sensitivity analysis was introduced. The sensitivity analysis tested how much the ranking of projects change when using alternative weighting schemes on the SEI scores for transport, water, and sanitation projects. Again, the weights assigned to variables in the SEI may be determined by PCA, or other methods, including equal weighting or a weighting scheme that reflects particular policy goals. For example, if there is an emphasis on extending services to the poor or creating employment, these factors in the SEI may be weighted more heavily. However, PCA is best when there is a preference to objectively assign weights. A prioritization scheme where decisions rely on subjective judgment may reduce legitimacy to government decisions, especially in environments with low institutional capacity. For transport, we tested two alternative weighting scenarios to compare outcomes with the project ranking by PCA-determined SEIs. These alternatives were equal weighting 7 and a weighting scheme that hypothesizes a government emphasis on most beneficiaries served and most jobs created. These alternative weights are summarized in Table 9. Table 9. Transport SEI Sensitivity Analysis BEN_COS EMP_COS POOR_COS PCA Weighting 0.638 0.439 0.633 Alternative scenario 1: Equal weighting 0.577 0.577 0.577 Alternative 2 0.650 0.650 0.394 Similarly, for water and sanitation projects (still evaluated separately), we tested alternative scenarios of equal weighting and ranking by the combined water-sanitation PCA weights. The latter (termed ‘Alternative 2’) can also be thought of as hypothetically representing a policy priority on overall beneficiaries and poor due to higher weights for these variables. These weighting scenarios are summarized in Tables 10 and 11. Table 10. Water SEI Sensitivity Analysis BEN EMP POOR PCA Weighting 0.638 0.439 0.633 Alternative scenario 1: Equal weighting 0.577 0.577 0.577 Alternative 2 0.687 0.233 0.688 Table 11. Sanitation SEI Sensitivity Analysis BEN EMP POOR PCA Weighting 0.598 0.516 0.613 Alternative scenario 1: Equal weighting 0.577 0.577 0.577 Alternative 2 0.687 0.233 0.688 To demonstrate the results of sensitivity analysis, we present three Infrastructure Prioritization Matrices each for transport, water, and sanitation. The results in Figures 14, 15, and 16 reveal an important finding, namely that project rankings change very little with alternative weighting scenarios. This suggests that SEI scores are much more sensitive to the values of input variables themselves that to the weighting system applied to combine them. It also demonstrates the robustness of results generated by the PCA-weighted Infrastructure Prioritization Framework. Nevertheless, alternative weighting may be still applied to meet policy requirements. 7. In keeping with PCA, .577 is used so that the sum of the square of each element equals 1. 22 Transport Sensitivity Analysis 100 P12 90 P15 80 P14 70 P1 60 P18 P16 P11 P4 50 SEI P13 Original IPM 40 P19 P10 30 P17 20 P9 P5 P7 P8 10 P2 P6 P3 0 0 10 20 30 40 50 60 70 80 90 100 FEI 100 P12 90 P15 80 P1 P14 70 60 P4 P18 P16 P11 Alternative 1 50 SEI P13 Equal weight 40 30 P19 P10 P17 20 P9 P7 P8 P5 10 P2 P6 P3 0 0 10 20 30 40 FEI 50 60 70 80 90 100 100 P12 90 P15 80 P14 70 P1 60 P16 P11 P18 Alternative 2 50 P4 SEI P13 40 P19 P10 30 P17 20 P9 P8 P5 P7 10 P2 P6 P3 0 0 10 20 30 40 50 60 70 80 90 100 FEI 23 Water Sensitivity Analysis 60 P15 50 P3 P10 40 P1 P7 P4 30 P12 P16 P11 P35 Original IPM P31 SEI P13 P5 20 P34 P32 P33 P6 P14 10 0 0 10 20 30 40 50 60 70 80 90 100 FEI 60 P15 50 P3 P10 40 P1 P7 P4 30 P12 P16 Alternative 1 P11 P31 P35 SEI Equal weight P13 20 P5 P34 P32 P6 P33 10 P14 0 0 10 20 30 40 50 60 70 80 90 100 FEI 60 50 P15 P3 40 P10 P1 30 P4 P16 Alternative 2 SEI P12 P7 20 P31 P35 P34 P13 P5 P32 P11 P6 10 P33 P14 0 0 10 20 30 40 50 60 70 80 90 100 FEI 24 Sanitation Sensitivity Analysis 100 P23 90 P24 80 70 P17 P27 60 P29 P22 P28 50 P21 P20 SEI P19 Original IPM 40 P18 P9 P25 P26 P8 30 20 P30 10 0 0 10 20 30 40 50 60 70 80 90 100 FEI 100 P23 90 P24 80 70 P17 P27 60 P29 P22 P28 50 P21 P20 P19 Alternative 1 P9 P18 SEI 40 Equal weight P26 P8 P25 30 20 P30 10 0 0 10 20 30 40 50 60 70 80 90 100 FEI 100 P23 90 P24 80 70 60 50 Alternative 2 P17 P27 P29 SEI 40 P28 P22 30 P20 P19 P21 P26 P18 20 P8 P25 P30 10 0 0 10 20 30 40 50 60 70 80 FEI 25 Step 5. Selecting Projects Projects categorized as ‘High Priority’ in Quadrant A are natural targets for project selection, as they represent the set of projects that score highly with respect to both social- environmental and economic-financial considerations. Projects in Quadrants B and C, on the other hand, score relatively high on either the SEI or FEI, but not both. If all Quadrant A projects are selected, the remaining budget could be allocated to either (a) all projects in Quadrant B, if social-environmental factors are privileged over financial-economic ones, (b) all projects in Quadrant C, if financial and economic factors are most important, or (c) a combined array of select projects from each. Selection of a mixed set from amongst Quadrants B and C makes space for expert review, flexibility, and informed political debate. The negotiated process of ordering projects within Quadrants B and C (and even those clustered around the quadrant intersection in Quadrant D) allows IPF to capture important information from the professional and political bases of knowledge amongst decision-makers. In other words, the framework informs decisions regarding projects in the medium-priority set, but leaves room for structured professional and political judgment. Considering Projects at the Intersection It is important to note that projects in Quadrant D are not automatically removed from consideration. While categorized as ‘Lower Priority’, they may be considered alongside High Priority FEI and SEI projects when they are clustered around the intersection of the budget line and also score relatively higher than some projects in Quadrants B or C along one dimension. This is demonstrated in Figure 10 above, where some Quadrant D water and sanitation projects clustered around the axis score relatively higher on SEI than some projects in Quadrant C, ‘High Priority FEI’. For example, Project 12, located in Quadrant D near the axis, has a higher SEI score than Projects 5 and 13 in Quadrant C, which have a higher FEI but low social-environmental rating. If government prioritizes social and environmental considerations above financial aspects, some projects in Quadrant D may compete strongly with those ‘High Priority FEI’ projects with relatively low SEI scores. Conversely, some ‘High Priority SEI’ projects score lower on FEI than do the ‘Lower Priority’ projects clustered near the axis. Tradeoffs: Considering High-Cost Projects In some cases, a project with a very high cost may fall within Quadrant B or C but score low on one dimension. As such, when considering the Quadrant B and C projects, implementation of the high cost project may mean that Government must forgo funding a number of other smaller projects. In this case, it would be recommendable to acknowledge tradeoffs between one large project and many others – a decision that comes with political ramifications. For example in Transport, Project 3 is positioned in Quadrant C, and is one of three very high cost projects (the others are Projects 2 and 6) with a total cost of $ 58,668,000. By foregoing Project 3, the Government could instead fund seven projects (Projects 1, 4, 5, 7, 8, 17, and 18) in Quadrant B, which all score higher with respect to SEI and total only $33,545,000 combined. 26 Figure 12. Considering Transport project tradeoffs, P2 versus all projects in Quadrant B 100 P12 90 P15 80 B P14 70 P1 P16 P11 60 P18 P4 50 SEI P13 40 P19 P10 30 P17 20 P9 P7 P8 P5 10 P2 P6 P3 0 0 10 20 30 40 50 60 70 80 90 100 FEI Next Steps: Implementing IPF The following section discusses challenges to IPF in Panama, key lessons drawn from the pilot, and next steps for full implementation. Challenges and Lessons The main challenge to overcome in moving towards full implementation of the IPF is the limited and deficient technical, financial and economic data at the project level. Calculation of the SEI and FEI depends on the availability and consistency of indicator values for each project. At present, these are missing for some projects or of undetermined accuracy. In 2001, the Public Investments System of Panama (SINIP) was created under the Panama Ministry of Economy and Finance (MEF) to store project-level information. One of the main institutional goals of the SINIP was to provide an information system for designing, evaluating, training and monitoring investment projects in the public sector (Executive Decree 148, December 11, 2001). While this could be a useful platform on which to base IPF, information consigned in the SINIP database is not currently sufficient to implement the framework. There are no established basic information requirements for all projects, and the definitions and consistency of variables stored vary from project to project, even within a particular sector. Moreover, this analysis has identified important data limitations in the selected sectors. For example, between 1995 and 2005, the Panama Ministry of Public Works (MOP) carried out economic evaluations based on the Highway Development and Management Model (HDM), versions 3 and 4. Technical and economic evaluations of road projects were based on the HDM. However, this practice was suspended 10 years ago, and since then, there is no systematic production of data for proper financial and economic analyses of transport projects. To strengthen the SINIP database and make it suitable for the IPF, we recommend adding a prioritization module to systematically gather a minimum set of relevant information to assess, 27 rank and select projects. This model would be used to populate the database on which to run the IPF analysis, as described in the steps above. Building this platform would require (a) a redesign or at least modification of the existing SINIP platform, and (b) a training plan to develop the institutional capacity within MEF to monitor the quality of data, interpret results and update the IPF platform. Three main lessons have emerged from this analysis. First, although a full CBA assessment for each of the proposed infrastructure projects to the MEF is unrealistic given the information requirements, set assumptions, and extensive technical and resource needs, we recognize the value of incorporating CBA elements on the financial and economic side of the IPF. To integrate available CBA elements into the framework, we calculated Benefit-Cost Ratios (BCRs) that contrasted the net present value of observable monetary (monetized) flow of revenues against the observable flow of costs per project. While this information significantly enriched the FEI estimations, it is still insufficient to provide a comprehensive picture of the financial and economic aspects of the infrastructure projects. For a full implementation of the IPF, we recommend to combine the BCR information with other financial and economic aspects, such as externalities, multiplier effects, and implementation risks. The second main lesson, drawn from sensitivity analysis, is that IPF is more sensitive to indicator values than the weights assigned to each variable. This finding shows that PCA weighting is a good approach to combining indicators into index values, as it is both objective and robust. The objectivity of index construction lends credibility to project selection and helps protect against over-politicization of project selection or allegations of particularistic policy making. The third lesson is that it may desirable to prioritize projects at the subsector level. In the separation of water and sanitation projects in the pilot, we see the benefit of prioritizing subsectors separately. This could also reasonably apply to the consideration of construction versus rehabilitation projects within a sector such as transport. The advisability of separating project analysis by subsector will depend on the number of projects to be prioritized, patterns in the results of combined analysis, and policy goals. To the first point, there should be a sufficient number of each sub-category of project to justify disaggregation. To the second point, if a certain category of project regularly scores comparatively low but is responding to a particular policy goal (e.g., sanitation projects), they may be considered separately. There are some additional needs related to variable definitions that must be dealt with moving forward. For one, variables must be carefully defined to ensure that PCA applies as intended. In the pilot, poverty levels (in percentage terms) were inversely related to other desirable factors (e.g., jobs created, beneficiaries). This resulted in PCA-determined weights that negatively weighted poverty, which stands at odds with policy goals. This may be dealt with in several ways, but our approach was to select a different poverty metric, the absolute number of poor beneficiaries, which both reflected poverty levels and also yielded positive PCA weights. Another variable decision, which may be made on a sector or subsector basis, is whether to choose indicators that are reflective of effectiveness or efficiency. For example, if the number of beneficiaries is known (as an input to the SEI), one could use the absolute value to help answer questions of effectiveness (e.g., expansion of service to the most people possible). On the other hand, an indicator such as ‘beneficiaries per dollar spent’ or ‘dollars per beneficiary’, would demonstrate a measure of efficiency. 28 Lastly, it is important for decision-makers to be aware that project outcomes tend to be better in high-income areas. In a past pilot application to Vietnam, we discovered an inherent bias towards infrastructure projects in wealthier regions attributable to better project preparation and generally better social-environmental indicator scores for direct beneficiaries or jobs created. Whilst the poverty score can counter-balance this, the issue is important to keep in mind. This can be further tested by mapping results to location and local economic and development conditions. Implementation Requirements Implementing the IPF to full infrastructure sectors will require the commitment of staff and resources for data collection and analysis, as well as the establishment of an institutional set- up and decision-making structure that can facilitate the IPF and translate its outputs into decision-making. In this section, we propose organizational and resource recommendations for implementation. Recalling the IPF process map, we identify three sets of requirements for each step: participants, capacities, and resources. I • Deliberation with decision makers, experts, and key Select Criteria stakeholders II • Source project data (CBA elements incorporated when available) Prepare Data • Transform / standardize data III Construct • Variables combined to create SEI and FEI Performance • PCA applied when objectivity is preferred Indices IV • Plot SEI and FEI coordinates Plot Visual • Quadrants defined by budget Interface constraint V • Based on informed political and technical debate Select Projects • Quadrant A = high priority Table 12 describes the general participant, capacity, and resource requirements needed for each step of the IPF process. Naturally, the set of participants included in each stage is subject to MEF guidance, but we make some preliminary suggestions here. 29 Table 12. IPF Implementation Requirements IPF Step Participants Capacities Resources Knowledge of key policy goals MEF Knowledge of financial and economic 1. Select criteria Sector agency project indicators at sector level Authority over project selection managers and experts Knowledge of available/possible social and environmental indicators Data: project proposals and constructed estimates, where Expertise in sector Consultants necessary 2. Gather / prepare Experience with basic economic and Line agency Project feasibility and pre- data financial calculations analysts feasibility study documents Fluency with Excel IPF platform (SINIP) database inputs Experience with basic economic / Consultants financial calculations 3. Construct MEF analysts STATA or other statistical software Fluency with MS Excel performance indices Line agency Populated project variable dataset Familiarity with STATA or other analysts statistical software Consultants Fluency with MS Excel 4. Plot visual MS Excel MEF Analysts and Familiarity with basic quantitative interface SEI / FEI index outputs sector experts analysis Project experience MEF Familiarity with political issues and Authority over project selection 5. Select projects Sector agency sectoral policy goals and budget allocation managers The table above proposes that the levels of authority required for successful implementation of each step are highest in the first and fifth steps, whereas technical analytical skills are highest in the second through fourth steps. The beginning (criteria selection) and final stage (project selection) of IPF are the most inherently political and require sufficient authority over infrastructure policy setting for IPF to be effective. In other words, the participants in Steps 1 and 5 must be actors who also have the authority to make investment decisions. The technical steps (2 through 4) may be relegated to technical analysts or consultants with sectoral experience and familiarity with quantitative analysis. The latter is important for checking the reliability and soundness of inputs sourced from project information sheets, and specialists may also be required to perform their own estimates of projected benefits, environmental impacts, and risk, depending on the data supplied by government. Improving Project Data There are two areas where project data may be improved: in the SINIP system and in pre- feasibility and feasibility studies. In consultation with MEF and line agencies, the SINIP can be made to include a prioritization platform that specifically requires the submission of IPF data during project proposal. The platform should include some data points that are generally applicable to all sectors, but may also be customized to reflect particular policy goals of a sector or subsector. For example, if the reduced congestion or improved connectivity are key transport policy goals, metrics can be designed for inclusion in the platform for transport projects. 30 Other data may be gathered from more detailed feasibility studies. At present, the proposed projects over 0.1% of the General State Budget require feasibility studies, and projects under this amount but over $5 million require pre-feasibility studies. The content of these studies and their basic requirements can be customized to meet the needs of implementing IPF. In particular, the feasibility studies must also be made to justify and explain the calculations or estimates of key social variables, such as number of beneficiaries and number of poor served, and important financial-economic inputs such as benefit-cost ratios, internal rates of return, or net present values. This requires better data on the streams of costs and revenues over a project’s lifecycle, as well as better projections of service, which are key to ensuring that the indicators submitted for each project are comparable across projects. In addition to improving data collection, index construction would be improved by better indicator specification. For example, the FEI would be strengthened by consideration of other important financial-economic factors, beyond the BCR, as well as by capturing other measures of benefits. For example, economic multiplier effects could be estimated with further disaggregation of national input-output tables by service sector for the construction industry. Better measures of implementation risk could help quantify the costs of potential implementation challenges. And the quantification of benefits could also be improved. To this last point, whereas transport estimates involved maintenance costs saved from rehabilitation and water estimates accounted for ‘benefits’ via tariff revenues, there was little data available to capture the benefits of sanitation projects, such as mitigation of health and productivity costs. Work done by the World Bank 8 on estimating the costs associated with under provision could be useful to inform estimation of these benefits. With respect to social and environmental indicators, measures of environmental impact (e.g., carbon footprint) may be greatly improved to reflect the environmental costs of proposed projects, and a measure of social risks should be better specified to capture potential human impacts. At present, a simple high-medium-low score was given to reflect these risks, which is too imprecise to be useful for IPF or decision-making, in general. These variable specifications need to be further developed in consultation with decision-makers. The pilot application of IPF demonstrates the analytical power of a systematic infrastructure investment decision framework and offers an approach to improve the evidence base for infrastructure decisions. The approach is objective and data-driven yet based on the application of consensus-determined decision criteria. This allows for transparency in investment decisions alongside improved responsiveness to policy goals. Furthermore, the use of evidence is balanced by a degree of flexibility to accommodate sector-level policy and incorporate valuable practice-based, professional, and political knowledge at multiple levels of government. 8. For an example methodology, see the 2008 Water and Sanitation Program’s Research Report ‘Economic Impacts of Sanitation in Southeast Asia’ which details their approach to estimating health, productivity, and water-related costs of sanitation. 31 References Araujo, J., Brueckner, M., Clavijo, M., Vostroknutova, E. and K. Wacker (2014): “Benchmarking the Determinants of Economic Growth in Latin America and the Caribbean,” World Bank Report No. 91015-LAC. Loayza N., Fajnzylbe P., Calderon C., (2004): “Economic Growth in Latin America and the Caribbean: Stylized facts, explanations, and forecasts”. Central Bank of Chile Working Paper 265. Marcelo, D., Mandri-Perrott, C., House, S., and Schwartz, J. (2016). An Alternative Approach to Project Selection: The Infrastructure Prioritization Framework. World Bank Working Paper, forthcoming. World Bank (2015). Panama: Locking in Success, a Systematic Country Diagnostic. 16 January 2015. World Bank b (2015). World Development Indicators. Retrieved August 2015 from http://data.worldbank.org/data-catalog/world-development-indicators. World Economic Forum (2014). The Global Competitiveness Report 2014-2015. Retrieved August 2015 from http://www.weforum.org/reports/global-competitiveness-report- 2014-2015. 32 Annex 1. Transport Projects SEI Inputs FEI Input ID COST Project Name BEN EMP POOR BCR SEI FEI Project 1 1,800,000 Rehabilitación Carretera de Anachucuna 6,733 33 5,992 0.242 68 8.2 Project 2 45,000,000 Rehabilitación Camino Tortí-Chimán-Gonzalo Vásquez 16,680 108 11,676 0.442 7 15.0 Project 3 58,668,000 Rehabilitación Carretera Las Tablas-Los Asientos-Cañas-Cacao 14,056 40 5,622 1.298 3 44.2 Project 4 2,997,000 Rehabilitación carretera de Irgandi (para llegar a la Comunidad de Playón Chico) 11,644 33 8,617 0.147 56 5.0 Project 5 4,800,000 Rehabilitación Carretera CPA-Calabacito 1,718 40 1,220 0.205 13 7.0 Project 6 63,000,000 Rehabilitación Carretera Circunvalación Garachiné-La Palma 6,604 130 4,557 0.214 3 7.3 Project 7 8,948,000 Rehabilitación Carretera Tambo-Las Marías y Construcción de Puentes Vehiculares (5) 5,240 42 4,559 0.253 13 8.6 Project 8 5,400,000 Rehabilitación Carretera Loma Del Naranjo-Tres Quebradas 831 40 391 0.405 9 13.8 Project 9 4,880,000 Rehabilitación Calles de Penonomé (Vista Hermosa) 3,063 35 1,011 0.994 12 33.8 Project 10 4,720,000 Rehabilitación Calles de Playa Leona (La Mitra hacia Peñas Blancas y Paso Arenas) 8,442 70 2,026 1.747 27 59.4 Project 11 1,600,000 Rehabilitación Calles de La Chorrera 1,354 46 555 1.191 38 40.5 Project 12 4,500,000 Rehabilitación Calles de Chilibre 53,955 60 18,884 1.321 100 44.9 Project 13 4,200,000 Rehabilitación Calles del Distrito de Aguadulce 12,881 58 2,834 0.614 33 20.9 Project 14 4,000,000 Rehabilitación Calles de Colon (Casco Antiguo y Arco Iris) 39,453 57 6,312 1.414 66 48.1 Project 15 2,000,000 Rehabilitación Calles del Distrito de Bugaba 17,698 46 3,540 2.939 74 100.0 Project 16 4,050,000 Rehabilitación Calles de Santiago (2015) 31,065 58 3,107 1.017 50 34.6 Project 17 8,000,000 Construcción Camino Quebrada Honda 6,923 66 3,877 0.161 17 5.5 Project 18 1,600,000 Rehabilitación Carretera Juay-Quebrada del Loro -Cañazas 1,616 44 1,503 0.227 44 7.7 Project 19 2,480,000 Rehabilitación Carretera Lagarterita-Los Hules-Las Pavas-Lagartera Grande y Mejora del Puente 2,996 30 1,438 0.464 24 15.8 33 Annex 2. Water and Sanitation Projects SEI Inputs FEI Input ID COST Project Name BEN EMP POOR BCR SEI FEI Project 1 17,650,597 Mejoramiento sistemas de agua potable y saneamiento- distrito de Colón 206481 125 74333 1.148 36.76 22.96 Project 2 16,194,901 Construcción línea paralela hacia Costa del Este 874935 138 166238 10.688 outlier Project 3 5,887,649 Construcción de línea de conducción del Tanque Ameglio - Red de 315019 156 63004 4.420 41.43 88.41 distribución del Sector 3 Project 4 5,991,593 Mejoramiento obras complementarias red de distribución - Área 67265 156 36996 0.495 22.93 9.90 Metropolitana (Línea de Conducción Pacora - Tanara - Tataré Project 5 7,450,110 Construcción de mejoras a la red de distribución de Chorrillo y Santa Ana 36512 98 13144 1.041 11.66 20.83 Project 6 5,971,133 Mejoramiento al Acueducto de San Francisco 43939 98 8348 1.231 10.85 24.62 Project 7 30,000,000 Construcción de línea de conducción de La Chorrera - Capira (Sectores 7 y 8) 53247 208 13844 0.494 20.12 9.87 Project 8 6,800,000 Construcción del sistema de alcantarillado sanitario de Parita 3723 191 1378 0.064 13.49 1.29 * Project 9 5,250,000 Construcción del Sistema de Alcantarillado Sanitario de Chilibre Centro 6813 189 1363 0.098 13.50 1.95 * Project 10 5,000,000 Construcción de línea paralela Sabanitas, Planta - Cuatro Altos 206481 153 74333 2.879 38.67 57.57 Project 11 15,300,000 Construcción de nueva planta potabilizadora de 10 MGD de Dolega y 16636 152 4658 0.138 12.28 2.76 sectores aledaños Project 12 60,000,000 Mejoramiento integral al abastecimiento de agua de David 116000 148 25520 0.532 21.97 10.64 Project 13 9,000,000 Mejoramiento al sistema de agua potable de Puerto Armuelles 47401 103 25123 0.603 15.45 12.07 Project 14 5,000,000 Mejoras al sistema de abastecimiento y distribución de agua potable de Santa 3905 102 1406 0.183 7.46 3.66 Marta, Santo Domingo Project 15 50,000,000 Estudio, diseño y construcción para una nueva planta potabilizadora de Río 156698 239 86184 0.596 44.96 11.92 Bayano Project 16 9,980,000 Construcción de obras complementarias para red de distribución de La 161470 103 41982 1.122 25.16 22.44 Chorrera (Estación de bombeo y proyectos complementarios La Chorrera) Project 17 40,000,000 Construcción Sistema de alcantarillado de Changuinola 31223 238 21544 0.101 22.96 2.02 * Project 18 5,000,000 Construcción de mejoras al Sistema de Alcantarillado Sanitario de Antón 9788 186 5579 0.110 14.47 2.19 * Project 19 10,000,000 Construcción del sistema de alcantarillado sanitario de Boquete 15339 208 4142 0.213 15.89 4.25 * 34 Project 20 8,000,000 Construcción del sistema de alcantarillado sanitario de Ocú 7006 203 1962 0.073 14.61 1.47 * Project 21 7,000,000 Construcción de alcantarillado sanitario de Macaracas 2890 203 1329 0.037 14.25 0.74 * Project 22 20,000,000 Diseño y Construcción Sistema de alcantarillado sanitario de Chepo 20420 234 11231 0.090 19.64 1.80 * Project 23 50,000,000 Diseño y construcción del sistema de alcantarillado sanitario de Arraiján 220779 243 50779 0.362 39.73 7.24 * Project 24 50,000,000 Mejora Integral del sistema de alcantarillado sanitario de La Chorrera 161470 243 41982 0.285 34.67 5.70 * Project 25 5,000,000 Diseño y construcción del sistema de alcantarillado sanitario de Capira 5181 186 2901 0.096 13.59 1.93 * Project 26 18,000,000 Construcción del sistema de alcantarillado sanitario de Bejuco – Chame 7980 184 3032 0.056 13.63 1.12 * Project 27 60,000,000 Construcción del sistema de alcantarillado sanitario de Santiago (Sistema de 64856 234 17511 0.221 23.35 4.42 * recolección y tratamiento de aguas residuales) CAF-II FASE Project 28 42,825,811 Ampliación y mejoras del sistema de alcantarillado sanitario de Chitré y 50694 210 10139 0.287 19.22 5.74 * alrededores - (Ejecuta el PAN) Project 29 24,397,000 Ampliación y mejoras al sistema de alcantarillado sanitario de Puerto 47401 208 25123 0.320 22.59 6.40 * Armuelles - CAF-II FASE Project 30 5,413,130 Construcción del sistema de alcantarillado sanitario de Metetí. 2313 118 1411 0.056 8.47 1.11 * Project 31 12,674,150 Construcción del sistema de abastecimiento de agua potable de Chorro 31537 162 11669 0.478 15.40 9.55 Blanco, Alanje - Boquerón - II Etapa Project 32 9,000,000 Captación y ampliación de la planta potabilizadora de Isla Colón - CAF-II FASE 7364 125 2651 0.109 9.50 2.18 Project 33 5,500,000 Mejoramiento al sistema de agua potable de El Valle de Antón 7602 98 2129 0.234 7.55 4.68 Project 34 5,000,000 Nueva toma de agua cruda en Bonyic, Changuinola 31223 103 21544 0.391 13.79 7.81 Project 35 5,000,000 Ampliación de la planta potabilizadora de Santiago 64856 148 17511 2.115 17.51 42.31 * Sanitation projects 35 Annex 3. Example Excel Project Data Form Project Prioritization Form Project Name Sector Total Cost (US $ dollars) Implementation time (Years) The empty fields below should be populated with the data you have gathered for each project. Once you click "Submit" this form will reset. thereafter you can enter in the data from your next project Social and Environmental Indicators (SEI) Financial and Economic Indicators (FEI) 1 Number of direct beneficiaries (BEN) 1 Internal Rate of Return (IRR) 2 Direct jobs during Implementation (EMP) 2 Economic Rate of Return (ERR) 3 Poor population targeted (POOR) 3 Net Present Value (NPV) 4 Region according poverty level (POV) 4 Benefit Cost Ratio (BCR) 5 Social and environmental risks (SER) 5 Multiplier effects (ME) 6 Carbon Footprint (CO2) 6 Implementation Risks (IR) In the following section please make a comment on how difficult you found it to gather the various pieces of data. If you made any adjustments to the recommended calculations, please make note of change in this section. Please provide any other info the World Bank may find helpful. t 36