Policy Research Working Paper 7674

Prioritizing Infrastructure Investment

A Framework for Government Decision Making

Darwin MarceloCledan Mandri-Perrott

Schuyler HouseJordan Schwartz

Public-Private Partnerships Cross-Cutting Solutions Area & Singapore Infrastructure and Urban Development HubMay 2016

WPS7674P

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

ed

Produced by the Research Support Team

Abstract

The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent.

Policy Research Working Paper 7674

This paper is a product of the Public-Private Partnerships Cross-Cutting Solutions Area and the Singapore Infrastructure and Urban Development Hub. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at dmarcelo@worldbank.org, cmandriperrott@worldbank.org, jschwartz3@worldbank.org, and shouse@worldbank.org.

Governments must decide how to allocate limited resources for infrastructure development, particularly since financing gaps have been projected for the coming decades. Social cost-benefit analysis provides sound project appraisal and, when systematically applied, a basis for prioritization. In some instances, however, capacity and resource limitations make extensive economic analyses across all projects unfea-sible in the immediate term. This paper responds to a need for expanding the available set of tools for project selec-tion by proposing an alternative prioritization approach that is systematic and feasible within the current resource means of government. The Infrastructure Prioritization Framework is a multi-criteria decision support tool that considers project outcomes along two dimensions, social-environmental and financial-economic. When large sets of small- to medium-sized projects are proposed, resources are limited, and basic project appraisal data (but not full

social cost-benefit analysis) are available, the Infrastructure Prioritization Framework can inform project selection by combining selection criteria into social-environmental and financial-economic indexes. These indexes are used to plot projects on a Cartesian plane, and the sector budget is imposed to create a project map for comparison along each dimension. The Infrastructure Prioritization Framework is structured to accommodate multiple policy objectives, attend to social and environmental factors, provide an intui-tive platform for displaying results, and take advantage of available data while promoting capacity building and data collection for more sophisticated appraisal methods and selection frameworks. Decision criteria, weighting, and sensitivity analysis should be decided and made trans-parent in advance of selection, and analysis should be made publicly available and open to third-party review.

PrioritizingInfrastructureInvestment:AFrameworkforGovernmentDecisionMaking

DarwinMarcelo,CledanMandri‐Perrott,SchuylerHouse,JordanSchwartz

Key words: policy planning, prioritization, infrastructure, decision support,multi‐criteriaanalysis,projectselection

JELclassificationcodes:H54,H76,R42,R53

2

Introduction

Infrastructureservices,widelydeemedcritical toeconomicdevelopment, tradeconnectivity, social welfare, and public health, are underprovided in manyregionsandtypicallyfeaturestronglyinnationaldevelopmentplans.Leadingupto2020,anestimatedUS$836billiontoUS$1trillionwillberequiredeachyearto meet growth targets worldwide (Ruiz‐Nuñez & Wei, 2015; World Bank).Global estimates of infrastructure investments required to support economicgrowth and human development lie in the range of US$65 trillion to US$70trillion by 2030 (OECD, 2006), while the estimated pool of available funds islimitedtoapproximatelyUS$45trillion(B20,2014).

Theseneedsareparticularlyintensifiedfordevelopingregions,asthechanginglandscapeofinvestmentandinternationalaidhasalsoreducedtheavailabilityofdonorfundsandshiftedthelocusofinfrastructuredecision‐makingfromdonorsto governments. Moreover, the past 20 years have witnessed a shift towardsdecentralized infrastructure planning and implementation in many countries.Subnational governments, regional entities, and sector agencies have beendelegatedresponsibilityforplanningandprojectselection,thoughaccountabilityforfundallocationsmayremainwiththecentralizedfinanceagency(CFA).Whilethese constituencies may propose numerous infrastructure projects,governments often have insufficient financial resources to implement the fullsuiteofproposals.Thisrequiresparingdownthesetsofproposedinfrastructureprojects,expandingthepoolofresources,orboth.

Good practice suggests that economic and strategic project appraisals andfeasibility studies provide a good basis for project prioritization via highestsocietalnetpresentvalue.Therealityformanycountries,however,isthattheylack the capacity and resources to provide extensive economic analysis acrossfull project sets or are challenged with having to make decisions based onincompleteorsecond‐bestinformation.Thus,thereisaneedforevidence‐basedinfrastructuredecisionsupport that isconsistent,pragmatic,andresponsive totheparticularneedsandcurrentcapacitiesofagovernment.

Thispaperbeginswithanoverviewofexistingapproachestoprojectselection,highlighting the challenges of prioritization. Next, we propose an alternativeapproachtoprioritization–theInfrastructurePrioritizationFramework(IPF)–thatutilizesexistingandaccessibledataviamulti‐criteriadecisionanalysis.TheIPF is intended to help governments systematically compare projects, whilepromoting the building of analytical capacity and data for more extensiveeconomicanalysis. Inthisway, it isanextensionofthesetoftoolsavailabletosupport project selection and is complementary to ongoing efforts to buildprojectappraisalandselectioncapability.TheIPF,asitispresentedherein,isthelatestversionofthetool,whichcontinuestoevolvethroughongoingpiloting.Wefollow its description with lessons drawn from initial pilots in Vietnam andPanama and conclude with a discussion of next steps to improving the tool’sapplicabilityandimplementation.

3

PrioritizingInfrastructure

Project selection implies grappling with the relative exigency, efficiency, andeffectiveness of investments. A number of steps areneeded to reachdecisionsthatmatch policy guidancewith project appraisal and subsequent investment.TheWorldBank’sUnifiedFrameworkonPublicInvestmentManagement(PIM),aligned generally with the Public Expenditure and Financial Accountability(PEFA) initiative, is a useful overarching framework for guiding governmentsthrough the processes of infrastructure investment and delivery, aimed atincreasingtheeffectivenessandefficiencyofinfrastructureinvestments.

The PIM framework identifies eight key “must‐have” features of an effectivepublic investment management system (see Figure 1). These constitute the“bare‐bones institutional features that would minimize major risks, beachievable in a lower‐capacity context, and yet provide an effective systemicprocess formanaging public investments” (Rajaram, Tuan, Bileska, & Brumby,2014,p.20).

In the PIM framework, project selection should follow first‐level screening,sound project appraisal, and independent review.1Where information andtechnical capacity is sufficient, governments may appraise projects via socialcost‐benefit analysis (SCBA) and extensive feasibility analysis. Thereafter,project selectionmay be based on selecting projects with highest net presentvalues(NPV),bestfitwithinfrastructurepolicyguidance,or(optimally)both.2

Figure1.KeyfeaturesofaPublicInvestmentManagementSystem

Source:PowerofPublicInvestmentManagement(Rajarametal.,2014)

Inpractice,however,manygovernmentsdonothavetheresourcesandcapacityneeded for generating extensive economic appraisals based on Social Cost‐Benefit Analysis and full‐fledged feasibility assessments across all proposedprojects.Inthesecases,thePIMapproach(Rajarametal.,2014,p.24)proposes

1. First‐levelscreeningshouldbedone toensure thatprojectsalignwith thedevelopmentstrategyand

meetbasicrequirementsforbudgetinclusionasaproject(Rajarametal.,2014).2. InChile,forexample,projectsareconsiderationbymultiplecriteria,includingNPVasakeyfactor.

4

that “the emphasis should be on basic elements of formal project appraisal,includingwhether

‐ Theneedforaprojectiswelljustified;‐ Theproject’sobjectivesareclearlyspecified;‐ Broadalternativeoptionstomeettheproject’sobjectivesareidentifiedand

comparativelyexamined;‐ Themostpromisingoptionissubjecttodetailedanalysis;‐ Projectcostsarefullyandaccuratelyestimated;and‐ Projectbenefitsareassessedqualitativelyaslikelytojustifythecosts.”

When governments must prioritize and select projects under conditions ofrestricted information and capacity, there is a risk that theymay fall back onunsystematic, ad hoc selection. In these cases, decision frameworks based onmulti‐criteriaanalysis canbehelpful to (a) systematizeprioritizationbasedonkeydevelopmentgoals;(b)makebestuseofavailable(orreasonablyattainable)information across the set of proposed projects; (c) invite governments to exante statedecisioncriteria tocontrolpropagationofwasteful “whiteelephant”projects; and (d) identify important missing information to improve projectappraisal and data collection looking forward. Such decision supportframeworks can help alleviate pervasive problems such as poor or reactiveplanning, regressive investment, over‐commitment, information asymmetries,corruption, and high degrees of political interference. We have undertaken areviewofthecommonprinciplesofandrationalesforsystematicinfrastructureprioritization,whicharediscussedinAnnex1.

Inresponsetoanobservedneedforanexpandedsetoftoolstosupportprojectprioritization and selection under conditions of imperfect or basic projectappraisal and limited resources, theWorld Bank developed an innovative andadaptable Infrastructure Prioritization Framework (IPF).3This paper discussesthe framework’s technical aspects and relevance to the broader investmentdecision process, along with lessons from the conceptualization of the tool inVietnamandfirstpilotapplication inPanama.Weconcludewithnextsteps forrefining the framework and exploring its applicability to identifyingopportunitiesforprivateinvestment.

AStepping‐StoneApproach:ImprovingProjectPrioritization

Recent attention to infrastructure prioritization is grounded on demonstratedgovernment and multilateral organization demand for evidence,comprehensiveness, value,and legitimacy in infrastructuredecision‐making.4Itis also a proposed precursor to identifying opportunities for private sectorinvestment.5Commonprinciplesofandrationales forsystematic infrastructure

3. See(Mandri‐Perrott,Marcelo,andHaddon,2014)foradiscussionofconceptualizationinVietnam.4. Inlate2014,theGroupofTwenty(G20)DevelopmentWorkingGroup(DWG)requestedMDBstotake

stepstoensurethatprojectpreparationfacilitiescollaboratetosupportgovernmentsininfrastructureprioritization. The draft ‘MDB’s Common Approach to Prioritizing Infrastructure with their PartnerCountries’ promotes a harmonized approach to project preparation, including use of standardizedenvironmental and social safeguards policies as well as common approaches including cost‐benefitanalysisandassessmentofprojectexecutability,developmenteffectiveness,andgreenhouseemissions(G20DWG,MDB’sCommonApproachtoPrioritizingInfrastructurewiththeirPartnerCountries,2015).

5. The 2014World Economic Forum (WEF) Investment Blueprint proposes that, “a strategic vision forinfrastructure should be the first step for a government to maximize investor financing in

5

prioritization are discussed in Annex 1. The provision of legitimate, evidence‐basedprioritizationisnecessarilyconstrained,however,byexistingcapacityandresourcelimitations.Manygovernmentsmakeinfrastructuredecisionswithonlybasic elements of project appraisal at hand. The challenge, therefore, is toprovide a provisional framing device that makes the best use of reasonablyattainable information in the immediate termuntil capacity and resources aresufficienttogeneratemoreextensiveeconomicappraisalsacrossfullprojectsetstosupportprioritizationdecisions.

In country consultations with Vietnam, Panama, Indonesia, and Peru, weobserved that local government units and line agencies proposed large sets ofprojects to the central government (e.g., the Ministry of Finance) for funding.Whileproposedprojectspassedpre‐screeningandwere,indeed,subjecttobasicappraisal, these did not always include full‐fledged SCBAor feasibility studies.Facedwith excessprojects, limited funds, andno single commondenominatorforcomparison,thecentralgovernmentsstruggledtomakesenseofthemassofproposals. Thus, therewas expressed demand for immediate decision supportwithin theexisting limitationsof the infrastructureplanningsystem,aswellasguidanceonimprovingdataforbetterprojectappraisalinthefuture.

For situations like this, we propose the IPF as a stopgap approach to projectprioritization that serves as an interim decision structuring tool until moresophisticatedpre‐selectionanalysesareavailable(seeFigure2).This‘stepping‐stone’approachdoes:

‐ Informdecision‐makingonprojectprioritization;‐ Compare projects that havepassed strategic pre‐screening and have been

subjecttobasicappraisal;‐ Makespacefortechnicaldeliberation;‐ Structure thedecision spacewhencapacityand information is limitedbut

neverthelesssufficientforsystematiccomparison;and‐ Encourage better appraisal by fostering discussion of key decision factors

forwhichprojectdatashouldbeimprovedorgatheredinthefuture.

Conversely,theapproachdoesnot:

‐ Deliveradefinitivelistofprojectsforselection;‐ Replacebestpractices inproject appraisal,particularlySocialCost‐Benefit

Analysis;or‐ Takecurrentdatadeficienciesasacceptableforthelongterm.

infrastructure.Thisvisionshoulddescribethegovernment’smediumtolong‐terminfrastructuregoals,alongwiththeunderlyingeconomicandsocialrationale,andenabletheprioritizationofapipelineofprojects in the shorter term” (2014,p. 19).AnearlierWEF report, ‘Strategic Infrastructure: Steps toPrioritise and Deliver Infrastructure Efficiently and Effectively’ (2012), proposes that governmentsmustdecidewhichsolutionscreatethegreatestimpactintermsofeconomicgrowth,whileconsideringsocialandenvironmentalissues.

6

Figure2.Visualizingthesteppingstoneapproachtoinfrastructureprioritization

Ad‐Hoc/UninformedProjectSelection

‐ Limitedproject‐levelinformationavailable

‐ Inconsistentuseofinformation

‐ Decisionsfrequentlybasedonnon‐technicalorpoliticalconsiderations

‐ Subjectiveassessmentcommon

SelectionInformedbyInfrastructurePrioritizationFramework

‐ Limitedinstitutionaland/ortechnicalcapacity

‐ Partialproject‐levelinformationavailable

‐ Projectcostsknown‐ Someinformationon

social,environmental,othereconomiceffects

‐ Decisionsbasedonminimumrelevantinformation

SelectionInformedbyAdvancedProject

Appraisal

‐ Hightechnicalandinstitutionalcapacityavailable

‐ Detailedproject‐levelinformationavailable

‐ Extensivequantifiedandmonetizedsocial,environmental,financialandeconomiceffectsknown

‐ Decisionsbasedonextensiveinformation

CurrentApproachestoPrioritization

PriortodescribingthemechanicsoftheInfrastructurePrioritizationFramework,we begin with an overview of approaches for project prioritization. Thisnaturallyrequiresdealingwithappraisal,which,whilenottheprimaryfocusofthispaper,influencestheapproachtoprioritizationandselection.6

Project comparison based on cost‐benefit analysis (CBA) allows projectcomparisonbasedona singlemetric –thatofmonetizedvalue.CBAessentiallytotals all costs and benefits of a project over its lifetime and discounts futureflowstocalculatepresentvalues.Akeystrengthisthatitallowsdecisionmakerstointuitivelycompareandrankdiversealternativesbasedonasingleindicator(Thomopoulos, Grant‐Muller, & Tight, 2009).With Social Cost‐Benefit Analysis(SCBA),prioritizationcanbebasedonselectingprojects thatmaximize thenetpresentvaluesforsocietyoverall.

Because SCBAassessments requirequantification andmonetizationofpositiveand negative effects, extensive information about the projects and theirprojectedimpactsisrequired(VanDelft&Nijkamp,1977).Sinceinformationinmanycontexts is limitedandmanycostsandbenefitsaredifficult tomonetize,andsinceSCBAitselfcanbequitecostly,thiscanmakethestandardatallorderfor application across all proposed projects. This is particularly difficultwhengovernments possess limited resources for appraising large sets of small‐ andmedium‐sized projects. An extensive discussion of CBA/SCBA can be found in

6. AgooddiscussionofthekeyelementsofsoundprojectappraisalmaybefoundinThePowerofPublic

InvestmentManagement(Rajaram,et.al,2014,p.76).

7

Annex 2, but most germane to this discussion are the resource, capacity, andtime limitations thatconstrainextensiveapplicationofSCBA inmanycontexts,imposingapracticallimitationontheappraisalmodeasabasisforprioritization.

SCBAisusedextensivelyintheUS,NewZealand,England,Australia,Singapore,Chile, Ireland, and many other countries to assess and prioritize alternativeinfrastructureprojects, particularly those that demand significant investments.But in the past five years, the UK, Australia, and many US states have alsopublished notes and guidance on the application of multi‐criteria decisionanalysis (MCDA), expanding the ‘Value for Money’ discourse to suggeststructured ways of employing MCDA to incorporate key policy criteria. Somecountries,suchasIreland,haveimposedthresholdstoguidewhengovernmentshould apply SCBA, multi‐criteria analysis, or more simple assessments,dependingonthesizeoftheproposedinvestment.

Appraisal and prioritization processes outside of OECD are largelyundocumented, but evidence suggests that prioritization is often based on apolitics, loose qualitative assessments, or professional judgment, but withoutclear principles underpinning selection (Petrie, 2010). More problematic, insomecontexts,prioritizationisnotbasedonformalappraisalatall,withprojectsapproved or disapproved on a rolling, ad hoc basis. The unstructured path toprojectapprovalinmanycountriesleavesroomforcorruption,inefficiency,andparticularist infrastructure policy that is unlikely to effectively servedevelopmentneeds.While countries strive to improve investment appraisal totheadvanceduseofSCBAacrosslargesetsofprojects,thereisaneedtosupportdecisions based onmore basic appraisal in the interim.MCDA canbe a usefulapproachtomakebestuseofavailable information,particularlydatareflectingkeycriteriadefinedbysectorornationalinfrastructuredevelopmentpolicy.

Multi‐CriteriaDecisionAnalysis

Multi‐criteria decision analysis has gained traction as a way of systematicallystructuring investment decisions when multiple aspects associated withproposed investments must be reconciled. Multi‐criteria decision approachesformalize the inclusion of non‐monetary and qualitative factors into decisionanalysisandcanbeusefulwheninformationoranalyticalresourcesarelimited.Indeed, MCDAs are currently included in government andmultilateral projectappraisalandselectionpracticeinregionsincludingthePacificIslandCountriesand Argentina, as well as in countries with longstanding and establishedprogramsofeconomicprojectassessment,includingChile,Ireland,andtheUK.7MCDAs have the added benefit of flexibility, since they can be recalibrated toaccommodateimproveddataasitbecomesavailable.

AnextensivediscussionofMCDAasappliedtoinfrastructuredecision‐makingisincluded in Annex 3, but we highlight some key points here. Practically,increaseduseofMCDAtosupport infrastructuredecisionsreflectspressureongovernmentstoworkwithintime, information,andcapacity limitations(DCLG,

7. See, for example, the work of the Pacific Island Infrastructure Facility (www.theprif.org) on

infrastructureselectionviaMCDM,UNFoodandAgricultureOrganizationworkonirrigationplanningin Argentina (http://www.fao.org/americas/eventos/vii‐taller‐irrigacion‐argentina/es/), or guidanceon multi‐criteria analysis for policy‐making (https://www.gov.uk/government/publications/multi‐criteria‐analysis‐manual‐for‐making‐government‐policy).

8

2009).Also,wheninformationisincompleteormultiplepolicygoalsareatstake,BeinatandNijkampsuggestthatthe“compromiseprinciple”ismoreappropriatethan the optimizing principle. This presumes a variety of decision criteria andstates that solutions must reflect a compromise between multiple priorities,while discrepancies between outcomes and goals are traded off by use ofpreferenceweights(Beinat&Nijkamp,1998).

In termsof intuitionand transparency,multi‐criteriamodels,which synthetizecriteriawithassignedweights, arepragmatic since theyare “able tocopewithalmostanyproblem”andareeasilyunderstood(Tsamboulas,2007). ImportantconsiderationsforapplyingMCDAtoissuesofsustainabledevelopmentincludetheabilitytodealwithcomplexsituations(criteria,differentscalesandaspects,type of data, uncertainties); the possibility to involvemore than one decision‐maker; and engagement of stakeholders to increase knowledge and proposealternativesolutions(DeMontisetal.,2004).

MCDAallowsfortwocriticalpolicychoices:(1)theselectionofcriteriabywhichalternativeswillbeassessed,and(2)theweightingofcriteria.TheseissuesarediscussedinmoredetailedinAnnex3,butinsummary,theselectionofcriteriaisessentialtocapturingthemostimportantcostsandexpectedimpactsofaproject,as well as performance with respect to prioritized development goals for thesector and country as a whole. Criteria weighting is also a policy choice.Weightingmaysimplybeuniform,whereinallcriteriaareequallyconsidered,orit may be subjectively set, with weights assigned via consultation or expertguidance to reflect the (expressed) relative importanceof thedecisioncriteria.Alternatively, weights may be statistically determined via methods such asPrincipal Component Analysis (PCA) to determine the linear combination ofcriteriathatcapturesmostofthevariationoftheunderlyingdata.

As with any approach, there are limitations and weaknesses associated withMCDA. For one, it lacks the utilitarian grounding in welfare economics thatcomeswithSCBA,whereinprojectselection isbasedonmaximizationofsocialwelfare (Layard & Glaister, 1994). There is also the threat of subjectivemanipulation of weights and criteria to privilege certain projects over others.Whiletheseweaknessesareapparent,weproposethreepointsinresponse.First,whilenotgroundedinutilitarianwelfareeconomics,MCDAisneatlyalignedwithextensive and well‐developed bodies of theory in policy analysis, democraticaccountability,anddeliberativegovernance,whereinpolicyselectionisbasedonthe stated goals of a polity and its citizens, including criteria of effectiveness,efficiency, feasibility, adequacy, equity, responsiveness, and appropriateness(Dunn, 2015; Araral et al., 2002;Weimer & Vining, 2015; Hajer &Waagenar,2003; Bardach & Patashnik, 2015; Poister, 1978).8 Second, the issue ofmanipulation to privilege the selection of particular projects is not unique toMCDA. Indeed, the threat of methodological corruption is present in everyapproach to appraisal and selection. Third, rules guiding implementation can

8. Responsiveness is of particular distinction, as it focuseson the extent towhich a policy satisfies the

needs, preferences, or values of the subjects of policy with societal standing (Dunn, 2015, 202).Moreover,asWeimerandViningpointout,“theappropriatenessofcost‐benefitanalysisasadecisionrule depends on whether efficiency is the only relevant value and the extent to which importantimpactscanbemonetized.Whenvaluesotherthanefficiencyarerelevant,cost‐benefitanalysiscanstillbeusefulasacomponentofmultigoalpolicyanalysis”(2005).

9

helpdealwiththelatterissue.Inparticular,decisioncriteria,criteriaweighting,andsensitivityanalysisshouldbedecidedandmadetransparent inadvanceofselection, and the data used and resulting analysis should be made publiclyavailableandopentothird‐partyreview.

TheInfrastructurePrioritizationFramework

TheInfrastructurePrioritizationFramework(IPF)isaquantitativemulti‐criteriaprioritizationapproachthatsynthetizesproject‐levelfinancial,economic,social,and environmental indicators into two indices – social‐environmental andfinancial‐economic–andconsidersthesealongsidethepublicbudgetconstraintforaparticularsector.The IPFdifferentiates fromothermulti‐criteriadecisiontools in fourways. First, it systematically incorporates policy goals, social andenvironmental sustainability considerations, and long‐term development aimsalongsidetraditionalfinancialfactors.Second,itispredicatedonparsimonyandpragmatism. Third, results are displayed graphically on an intuitive, graphicalinterface by which decision‐makers can compare alternative investmentscenarios. Fourth, it facilitates active deliberation of key decision criteria andpriorities for improving project appraisal looking forward. In this way, theprocessitselfisasimportantasitsoutputs.

SeveralempiricalissuesmotivatedtheconstructionandongoingdevelopmentofIPF. For one, there are significant challenges facing many governments ininfrastructure planning, wherein large numbers of infrastructure projects areidentifiedindevelopmentplans,whicharetobeimplementedwithscarcepublicresources, limited institutional capacity, andcostand timeconstraints. Second,these difficult decisions must be made based on currently available orreasonably attainable information for the set of projects. Third, given theimperfect nature of appraisal, it is important that projects be evaluated for“social(includingenvironmental)andeconomicvalue”(Dabla‐Norrisetal.,2011)in addition to financial impacts, but many of these social and environmentalimpactsaredifficult tomonetize.Fourth, there isadesire tobalanceanalyticalefficiency,derivedfromstandardization,withpolicyandpoliticalresponsiveness,derivedfromselectionofdecisioncriteria.

To the last point, IPF recognizes that the selection of infrastructure projectscannot be divorced from the political economy of project selection. Particularprojectsmay be chiefly valued by governments and other stakeholders due tokeypolicygoalswhicharenon‐economicinnatureorduetoconsiderationsthatobjective indicators cannot measure, such as promoting social cohesion,honoring culture, or redistributing wealth to the poor. But selection can beresponsivetopolicywithoutbecomingaltogetherpolitical.Assuch,thissupportframeworkexplicitlyaccommodatespolicyresponsivenessintwoways:throughcriteriaselectionandbyleavingadegreeoffreedomindecision‐makingthroughmultiplereferencesforjudgment(i.e.,twoindices).

In addition to building space for political deliberation, consultation, andprofessionaljudgment,thefollowingdesignidealswereincorporated,basedonasurveyofinternationalbestpractice:

‐ Thestrategicrelevanceofaprojectmustbedeterminedatthesectorlevelaswellaswithintheappropriatetierofgovernment;

10

‐ Project comparison should be systematic and based on quantitativemeasures,tothegreatestextentpossible,inordertolimitsubjectivity;

‐ Standardindicatorsofsocialvalueandfinancialreturnshoulddriveprojectcomparisons;and

‐ Theoutputshouldbetransparent,allowingforaclearaudittrail.

A key strength of IPF is that it may be flexibly applied. The framework canincorporateelementsfromothercommonmethods,suchasexpertjudgmentandcost‐benefit analysis. Expert judgment and deliberation come into play via theselectionanddefinitionofcriteria,aswellas intheselectionofprojectswithinthebudget constraint. IPF canalso takeadvantageof financialorpartial socialCBAcomponentsthataremoreeasilyquantified,measured,andmonetized(e.g.,net present values of market‐based costs and revenues). Nevertheless, IPF’smostimportantvalue‐addisinrelievingsomeoftheburdenofdeterminingandjustifyingtheassumptionsrequiredtomonetizeallbenefitsandcosts.

TechnicalFeatures

The IPF is designed to account for the inherent multidimensionality ininfrastructure planning processes. Stakeholders determine specific projectindicators or ‘selection criteria’ via a consultative process. Criteria may differfrom country to country and sector to sector. For example, the financial andeconomicindexmayincludeindicatorssuchasmultipliereffectsandnetpresentvalues, while the social and environmental index may include the number ofbeneficiaries,carbonfootprintandjobscreated.

The IPF aggregates criteria into two composite indices – social‐environmentalandfinancial‐economic–viaaweightedadditivemodel.9Inpractice,informationmay be quantitatively or qualitatively recorded, depending on the attribute.Qualitativedataiscommonwhenassessingsocial‐relatedphenomenaandhasanimportant informative value in infrastructure projects. To make use of thisinformation,amethodtoconvertqualitativeinformationintonumericaldataisapplied.

The twocomposite indicesallow for thecomparisonofprojectsagainstotherswithin a sector. The key output is a graphical display of projects’ relativeperformance within the sector under study along two axes, defined by thefinancial‐economic and social‐environmental composite index scores. Projectscores are mapped onto a Cartesian plane, whereby alternative investmentscenariosforasectormaybeconsidered.

The available budget for the sector is considered as a fixed amount andsuperimposed upon each of the axes. As a result, the budget constraint setsquadrantsontheplane(seeFigure5).FuturedevelopmentsoftheIPFmaytestalternativeapproachestocapturerelationshipsbetweenthetwoindicesandthebudget constraint (e.g., a sloped single budget line) or to account for privateparticipation through PPPs and other private financing schemes,whichwould

9. Acompositevalueorindexvalue isasinglenumericalfigurethatcombinesinformationfromseveral

underlying variables. The strength of this approach is that a decision–maker canefficiently considercomplexphenomenalikeeconomicperformance,sustainability,orcompetitivenessinasinglevariable(Freudenberg,2003;Nardoetal.,2005).AselectionofwidelycitedindicesislistedinAnnex4.

11

make the budget constraint variable. These technical features are furtherdescribedinthefollowingsections,whichdescribeIPFstep‐by‐step.

ConstructingtheIPFIndices

Thefirststepof IPFisto identifythesetof indicatorsthatwillbecombinedtoconstructthesocial‐environmentalandfinancial‐economicindices.Theselectionofvariables–orprojectcomparisonindicators–maydifferamongstapplicationcontexts based on government policy goals (e.g., particular sectoral, economic,social, and environmental aims) and stakeholder consultation. For explanatorypurposes,however,wewilldiscuss the IPF indexconstructionasperformed inthefirstconceptualizationofIPFinVietnamandinitialpilottestinPanama.10

The selection of variables seeks to preserve the principle of parsimony.Accordingly, this methodology relies on and requires a minimum level ofrelevantinformationtocompareexpectedoutcomesofalternativeinfrastructureinvestments. Multiple variables are selected to reflect different aspects ofexpected performance in two composite indicators, the Social andEnvironmental Index (SEI) and the Financial and Economic Index (FEI), eachbuilt on quantitative and transformed qualitative variables combined via anadditivemodel.11

Tocondensedissimilardatatypesandscalesofmeasurementintoindices,threedatatransformationsarerequired.Onemust(a)transformqualitativedataandordinalquantitativedataintousablescalardata,whereintheintervalsbetweenvaluesreflectdegreesofdifference;(b)standardizecriteriameasurementstoacommonscale;and(c)establishweightsforeachcriterionintheadditivemodel.

The transformationofcategoricalandordinalqualitativeandquantitativedatainto usable numerical data may be done using the Alternating Least SquaresOptimalScaling(ALSOS)algorithm,awidelyacceptedtransformationapproach.Within a quantified categorical variable, the numbers assigned by the ALSOSalgorithmtoeachcategoryreflectthedistancebetweencategories,revealingtheimplicitmetricofthevariable(Perreault&Young,1976).12

Numerical values are thereafter standardized via a typical standardizationformulatotransformthemeasurementstohaveameanvalueofzeroandunitvariance.Thestandardscore ofarawscore is

(1)

where isthesamplemeanand isthestandarddeviationofthevariable jforprojecti.

10. The work in Vietnam is described in detail in Mandri‐Perrott, Marcelo, and Haddon, 2014. At the

request of Vietnam'sMinistry of Planning and Investment (MPI), theWorld Bank piloted the IPF toprioritizeandselectpublicinfrastructureinvestments.OneoftheobjectivesoftheIPFinthiscasewasto provide the Government of Vietnam (GoV) with the means to operationalize the guidelines andrequirementsofthePublicInvestmentLaw(PIML)inanopenandtransparentmanner.ThePilotTestcoveredoverprojectsinthreesectors:transport,irrigationandurban.

11. To generate a valid numerical expression for the composite indices, it is required that qualitativeinformationistransformedintoquantitativevariables.

12. Statistical software such as SPSS and SAS include routines to easily perform transformations ofqualitativevariablesintonumericalvariables.

12

Once qualitative information is transformed and standardized to isolate thevariousunitsofmeasure,itiscombinedincompositeindices.Thestandardizedindicators are multiplied by weights to create index scores. Weights may beselected by decision‐makers or determined by statistical methods. Subjectiveweightingshouldbebasedondeliberationwithkeystakeholders toreflect therelative importance of component indicators. The strength of this weightingapproachisinstructuringdiscussionsontherelativeimportanceofcomponentindicators and policy goals. The risk, however, is in exposing the process tomanipulation in pursuit of prevailing interests. As such, regardless of theweighting systememployed, it is important thatdecision‐makers consider anddecide the weighting, in advance of analysis, and record these decisionstransparently.

In the cases of Vietnam and Panama, however, a statistical method, PrincipalComponentAnalysis(PCA)wasusedtodeterminetheweightsofeachvariableintheindex’sadditivefunction.OneofthemaincharacteristicsofPCAistheabilityto calculate coefficients based solely on the statistical relationship betweenvariables.Thisisusefulwhenthereisanexplicitpreferencetoobjectivelyassignweights and likelihood of redundancies in underlying data. In Vietnam, theMinistry of Planning and Investment expressed a desire to eliminate allsubjectivityfromtheprocess,whereastheusePCAinPanamawaspracticalandexperimental,basedonthedesiretotesttheapproach. Annex6includesmorenotesonPCA.

SocialandEnvironmentalIndex(SEI)

Infrastructureprojectsaremeantto improvequalityof life.Anumberofdirectsocial and environmental benefits are associated with their implementation,including improvedaccesstopublicservicesand joband incomeopportunitiescreated during the construction and execution of investments. These benefitscomeatacost,however.Engineeringworksfrequentlyrequireclearingforestedareas, polluting and endangering natural environments, and sometimesconstructionworksinvolvetheresettlementoffamiliesorcommunities.TheIPFdirectly considers relevant social andenvironmentalbenefits and costs via thesocialandenvironmentalindex(SEI),whosesub‐componentsaredependentontheevaluatinggovernment’sselectedcriteria.

InVietnam,theSEIconsistedoffiveindicators:DirectJobsCreated(DJ);NumberofDirectBeneficiaries(NB);PeopleAffectedbyRepurposingofLandUse(PA);Cultural and Environmental Risks (CER), and Pollution, in terms of CO2equivalent emissions (CO2). The data required to compute each variablewereprimarily sourced fromexistingproject feasibility studies.Additional variablesonprojectedindirecteffectswereestimatedusingdataroutinelygatheredbytheNationalStatisticOfficeofVietnam(Marcelo,Mandri‐Perrott,&Haddon,2015).13PCAwasusedtosynthetizesocialandenvironmentalvariablesintoacompositesocial‐environmental index. Since many social and environmental data were

13. The information collected was used to populate a Microsoft Excel form, which in turn populated a

databasewith embeddedmacros, programmed to transform thedatausing transformation functionsandweightsrequiredtocalculatethetotalSEIscoreforaproject.

13

recorded as qualitative ordinal variables, they were transformed using theALSOS14methodology.

Finally, PCA was used to generate an index function from the first principalcomponent, where the coefficients that maximize the variance of the firstprincipalcomponentareusedastheweightsforeachstandardizedvariable.TheresultantSEIfunctionwasexpressedasfollows:

(2) 2

ThelinearcombinationofthestandardizedvariablesZDJ,ZNB,ZPA,ZQCERandZCO2isequivalenttotheSEIadditivefunction.Thecoefficients , … , canbeinterpreted as the relative weights of SEI variables. Had weights beensubjectively determined or set to be equal, these would replace the PCA‐determinedcoefficients( values)above.

The resultant SIE calculations are normalized and rescaled to generate scoresbetween0and100foreachproject.Therescaled scoreis

(3) 100

where istheminimumvalueforvariableZand isthemaximum.

In Panama, on the other hand, the SEI consisted of only three indicators: thenumberofbeneficiaries, thedirectnumberof jobs created, and thenumberofservice recipients living below the poverty line,whichwere combined via thesameapproach.

FinancialandEconomicIndex(FEI)

The same procedure was used to construct the FEI, only with differentcomponentvariables.Financialprofitabilityandeconomicvalueareprobablythemostcommoninvestmentdecisionconsiderations.Thatsaid,publicinvestmentdecisions must also consider externalities and may include indirect economiceffects,suchasmultiplierandnetworkeffectsonotherindustriesandeconomicsectors.TheFinancialandEconomicIndex(FEI)seekstocondensetheminimumamountofrelevantinformationrequiredtoappropriatelyrepresentthefinancialandeconomiceffectsderivedfrominfrastructureinvestments.

In the case of Vietnam, the FEI consisted of five indicators selected by thegovernment in consultation with sector specialists: Financial Internal Rate ofReturn (IRR),Multiplier Effects (ME) determined by an Input‐Outputmodel, acategorical score indicating theproject’s locus indesignatedPriorityEconomicZones(PEZ),aqualitativemeasureofImplementationRisk(IR),andaqualitativemeasure of Complementarity/Competition effects (CC) intended to reflect thedegreeofalignmentofeachprojectwithexistinginfrastructure.

AswiththeSEI,datarequiredtocomputeeachvariablefortheFEIweredrawnfromfeasibilitystudies(FSs).Additionaldataonindirecteffectsgeneratedcouldbe calculated using input‐output table data collected by theNational StatisticsOffice.

14. Forexample,culturalheritagelossandenvironmentalriskswerequalitativelyclassifiedfrom"high"to

"low".ToapplyPCA,thevariablewastransformedtoaquantitativevalue.

14

AFEIwascalculatedusingPCA,asfollows:

(4)

where the suffix Z again denotes standardization, the suffix Q indicatesquantificationofaqualitativevariable,andthecoefficientsδ11,…,δ15arePCA‐determined weights. As with the SIE, the resultant FIE calculations werenormalizedandrescaled.

In Panama, on the other hand, the FEI required only standardization but notadditive combination, sinceonly onebase indicatorwasused.Data limitationsweresuchthatonlyoneindicatorwasavailable:thebenefitcostratioemergingfromfinancialcost‐benefitanalysis(andpartialsocialCBAintransport).

ComparingProjectsbySEIandFEI

Constructionof theSEIand theFEI composite indicatorsallows the rankingofprojectswithinasector,accordingtoprojectedrelativeperformancealongeachdimension. But a good infrastructure investment, in terms of financial andeconomicperformance,maysimultaneouslybeapoorchoice fromasocialandenvironmentalperspective,andviceversa.Thus,policymakersshouldnotmakedefinitive investment decisions based on only one dimension. In fact, neithershoulddecisionsbemadeonboth,withoutthe inclusionofacriticaladditionalinformationpiece–thepublicbudgetconstraint.

Projects are first plotted on a two‐dimensional Cartesian plane, with axesdefinedby the SEI andFEI scores. In Figures5 and6, eachpoint represents aproposedinfrastructureproject,withinonesectoronly.Thelocationofaprojectintheplaneisdeterminedbycoordinates(x,y)definedbythe(FEI,SEI)pair.

Comparison by each dimension does not presuppose that the FEI and SEIdimensionsareequallyimportant.Rather,itdoesquitetheopposite:itconcedesthattherelativeimportanceofSEIversusFEIisunknownandmakesnoattempttocombinethetwointoasingleindex.Whilefinancialandeconomicmeasuresmayremainthemostimportanttoagovernment,theSEIgivesdecision‐makersboth a point of reference for understanding development impacts and alsofacilitatestheidentificationofpotentialproblemsthatmayariseinthesocialandenvironmentaldimension.Thiscanhelpgovernmentsidentifycasesthatrequirefurthermitigationattentiontomaketheprojectviable.

TheBudgetConstraint

Onceprojectsareplotted,thebudgetconstraintisconsideredandsuperimposedseparatelyforbothdimensions,perpendiculartoeachaxis.Thebudgetlimitmaybebasedonaknownfixedamountoronanestimationbasedonthehistoricalproportion of funding requested actually allocated. It is important that agovernment’s fiscal and budgetary framework establish envelopes for publicinvestmentinordertosupportasustainableinvestmentprogram(Rajarametal.,2014,p.21).InVietnamandPanama,historicalbudgetallocationswereusedtoestimate budget envelopes for the following strategic five‐year investmentperiod.

Tolocatethepointofintersectionwherethebudgetconstraintcrosseseachaxis,projectsare first rankedaccording to theSEIandFEI scores.Then, thebudgetlimitisplacedatthepointwherethelastprioritizedprojectalongeachaxiscan

15

be funded. For example, in Panama, the total cost of proposed water andsanitationinfrastructureprojectswasUS$622,091,173.BasedonrecommendedfiguresfromtheMinistryofEconomicsandFinance, in turnbasedontheDraftAnnualBudget,thebudgetlimitforwaterandsanitationprojectswasestimatedtobearound55%ofthetotalcostofproposedprojects.Consideringthecostsofthetop‐rankingSEIprojects15,3,23,andsoon,thebudgetwouldbeexhaustedafter Project 29, at which the cumulative expenditurewould be $318,906,839(SeeFigure3). Since Project 12 costs $60,000,000, therewould be insufficientfundingtoincludeit.

Figure3.SEI‐prioritizedwaterandsanitationprojectswithinthebudgetlimit

The samewas done for the FEI (Figure 4). For water and sanitation, the lastprojectthatcouldbefundedwasalsoProject29,withacumulativeexpenditureof $304,002,232. The cumulative totals are different because the ranking ofprojectsaccordingtoeachindexaredifferent.

Figure4.FEI‐orderedwaterandsanitationprojectswithinthebudgetlimit

VisualizingtheSEI,FEIandBudgetLimitSimultaneously

SEI and FEI scores are plotted in a Cartesian plane. The budget constraint isimposedonto theplane following the logicdescribedabove. Since this isdonealong each axis, rather than delineating a singular threshold, the budgetconstraintresultsinquadrants(seeFigures5and6).Whileafunctioncouldbedetermined to express the relationship of SEI to FEI, decision makers are

0

10

20

30

40

50

60

Project 15

Project 3

Project 23

Project 10

Project 1

Project 24

Project 16

Project 27

Project 17

Project 4

Project 29

Project 12

Project 7

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 9

Project 8

Project 11

Project 5

Project 6

Project 32

Project 30

Project 33

Project  14

budget lim

it

0

10

20

30

40

50

60

70

80

90

100

Project 3

Project 10

Project 35

Project 6

Project 1

Project 16

Project 5

Project 13

Project 15

Project 12

Project 4

Project 7

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 9

Project 25

Project 22

Project 20

Project 8

Project 26

Project 30

Project 21

budget lim

it

16

unlikely toknowor come toeasyagreementon the relative importanceof thetwodimensions,particularlysincetheirmeaningisabstracted.15

Projects that fall inside the budget constraint along each axis represent the‘Investment Possibilities’ set within each dimension. For example, from a FEIpoint of view (X axis), the location of the budget constraint line indicates thethresholdwhere public resourceswould be fully exhausted. In the example ofVietnamtransport, resourceswouldbesufficient to financeonlythoseprojectswithaFEIabove70.FromaSEIperspective,ontheotherhand,resourceswouldbeenoughtofinanceonlythoseprojectswithascoreabove46.

Figure5.PrioritizationMatrix,VietnamTransportProjects

Theprioritizationmatrix forPanamawaterand sanitationprojects shows thatprojects with an FEI score above 6 and an SEI score above 22 would beconsideredashigherpriorityduetogoodperformancealongbothaxes.

15. Future developments to establish the relationship between SEI and FEI could be used to trace a

functionforthebudgedconstraint.Atthisstageitisassumedasafixednumber.

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100

SEI

FEI

B A

CD

A Higher priorityB Higher social/environmental

priorityC Higher financial/economic

priorityD Lower priority

17

Figure6.PrioritizationMatrix,PanamaWaterandSanitationProjects

InterpretingtheQuadrants

Quadrant A contains high–priority infrastructure projects that simultaneouslyscorehighontheSEIandFEI(greenpoints,Figures5and6).Theseprojectsarerecommended for implementation. On the other hand, projects falling intoquadrant D (red points) may be classified as lower–priority, since they scorerelativelylowonboththeSEIandFEI(reddots,Figures5and6).

Projects in quadrants B and C have two common features. First, they scorerelativelyhighoneither the SEIorFEI, butnotboth. Second, all ofprojects ineitherquadrantBorC,oracombinedarrayofselectprojectswithineach,couldbe implementedwithpublic funds. If theSEI isdefinitivelyprivilegedover theFEI, all projects inquadrantB couldbe selected for funding. Conversely, if theFEI is unequivocally more important to a government, all of the projects inquadrantCcouldbeimplemented.Alternatively,someportionofquadrantBandquadrant C projects could be funded,where both the FEI and SEI are deemedimportant.QuadrantBandCprojectsmaybegivenamediumprioritylevel.

Identificationofthesemedium‐priorityprojects leavesspaceforexpertreview,flexibility, and informed political debate. Since only a set of projects must beselectedfromamongstthese,thenegotiatedprocessoforderingprojectswithinQuadrants B and C allows IPF to capture important information from theprofessional and political bases of knowledge. In otherwords, the frameworkinformsdecisionsregardingprojectsinthemedium‐priorityset,butleavesroomforstructuredprofessionalandpoliticaljudgment.

Financial and economic considerationswill likely remain of key importance toprojectselection;however,itisrecommendedthatdecisionmakersdiscussanddocument theprinciplesofproject selection in advanceof results.Theex‐ante

P3

P4

P5

P6

P7

P10

P11

P12

P13

P14

P15

P16P17

P18

P19

P20

P21

P22

P23

P24

P25P26

P27

P28

P29

P30

P31

P32

P33

P34

P35

P1

0

5

10

15

20

25

30

35

40

45

50

0 10 20 30 40 50 60 70 80 90 100

SEI

FEI

AB

C

D

18

discussion and agreement onguiding principles is helpful toprevent decision‐makersfromsimplycherry‐pickingfromamongstremainingprojects.

One way to guide selection from amongst Quadrants B and C would be toestablishminimumacceptable threshold scores for individual variableswithineachcompositeindicator.SinceSEIandFEIscoresthemselvesdonotrepresentameaningfulquantifyorperformancescore(i.e.,theyholdmeaningonlyasscoresof performance relative to other proposed projects), it would be arbitrary toassignbasicthresholdsfortheindicesoverall.Butcomponentindicatorsmaybeusedtosetbasicrequirements.Forexample,projectsmayberequiredtomeetaminimum cost‐benefit ratio, extend services to areaswith aminimumpovertyprofile,ornotexceedasetcarbonemissionslimit.

LessonsfromVietnamandPanamaPilots

The early construction of the IPF in Vietnam and pilot application of the inPanamaweremetwith interest from stakeholders in both countries (Mandri‐Perrott,Marcelo,&Haddon,2014;Marceloetal.,2015).Theexperiencesbroughtto light a numberof issues for further refinement to the framework, however,whicharediscussedinthissection.Bothgovernmentswereamenabletoemployamulti‐criteriaprioritizationapproach,giventheirrespectivestrategicplanningcycles, infrastructure needs, fiscal plans, and legislative and governmentalsupportforemployingaprioritizationmethodology.Whilethetwoexperiencesareanimportantstart,moreworkisneededtoexploreimplicationsofregionaldisparitiesinthecontextofapplicationanddataavailability,and indevelopingspecificlinkstoprojectfinancingandPPPidentification.

VietnamExperience

InVietnam,twofactorsmadeprioritizationanaturalpursuitfortheMinistryofPlanning and Investment. First, the 2014 Public Investment Law referredspecifically to implementing a classification and selection system for proposedinfrastructure, to incorporate assessments of financial efficiency andeffectivenessalongsidesocialandenvironmentalsustainability.Second,thepilottestalignedwiththegovernment’sstrategicplanningcycle.Theexercisecovered30randomlyselectedprojectsinthreesectors:transport,irrigation,andurban.Animportantpre‐conditionwasthatprojectsshouldhavealreadyundergoneapreliminaryfeasibilitystudy(FS)(Mandri‐Perrott,Marcelo,&Haddon,2015).

Followinganinitialexploratoryandconsultationperiod,theIPFwasemployedforexanteprojectevaluationinthreesectors:transport,irrigation,andurban,ina two‐stageapproach.SincetheMinistryofPlanningandInvestment(MPI)didnot have sufficient funding to support a full set of feasibility studies for theapproximately3,000proposedprojects, an initial qualitativeproject validationand classification filter was applied to identify a subset of projects for whichfeasibilitystudiescouldbefunded.16Followingthefirstfilter,about268projects

16. Thefirstfilterreducedthesetofprojectsunderconsiderationbyassessingwhethertheymettestsof

(a) legal and procedural validity; (b) strategic validity (alignment with development goals); and (c)financial validity (capital valuewithin available resources). Projects that passed all three testswerethereafterclassifiedgeographicallytoidentifythoseprojectslocatedineither(a)prioritydevelopmentareasduetopovertylevelsorkeyeconomicdevelopmentinitiativesor(b)environmentallyprotectedregions. Naturally, the former were assigned higher classification scores, whereas the latter were

19

wereselectedforfeasibilitystudygrants.Ofthese,thirtyprojectswererandomlyselectedtorunthepilot:teneachfromtransport, irrigation,andurbansectors.The calculation of the SEI and FEI and the plotting of the projects against thebudgetconstraintfollowedthestepsdescribedabove.

TheprimarylessonsdrawnfromexperiencesinVietnamaresummarizedhere.First, there are important pre‐analytical steps required to ensure sufficientlycomparable data. One of the challenges of the pilot was that some data, evenfromwithinfeasibilitystudies,waseitheropaquelydetermined(e.g.,IRR)orhadlimited comparability across projects (Mandri‐Perrott, Marcelo, and Haddon,2015).Feasibilitystudiesshouldfollowclearrules,guidelines,andstandardstoensurequalityandcomparabilityofdata.

Second,specialattentionmustbegiventothemetricsusedtomeasurevariablesifPCAisusedtoassignweights.SincePCAsynthetizesinformationbasedonthecorrelations between variables, it is important to make sure that the weightsreflect the expected relationship between the variables and the compositeindicator. In some cases, this can be done with alternative specifications of acomponentvariable.Forexample, theconceptofpovertymaybeexpressedbythe population of poor residents or the poverty ratio. Because these decisionsare technical and require a certain degree of statistical and methodologicalknowledge,itisimportantthatdecisionsonvariablespecification–particularlyifa statisticalmethod such as PCA is proposed– aremade by professionalswithsufficientknowledgetodoso.

Third, pre‐filters or additional variables may be required when there areinherentbiasesinthesetofprojectsproposed,orwherethegovernmentaimstobreakregressivepatterns.Forexample,inVietnam,itwasobservedthatprojectsinpoorerregionstendedtoscoreloweronsomevariableinputstotheFEIorSEI.Thisobservation justifieduseofan initial filter toreflectagoal to targetareaswithhigherpovertyrates(Mandri‐Perrott,Marcelo,&Haddon,2015).

Fourth, in order to improve robustness of results and foster concurrentapplication with other supportive analytical tools (including CBA and expertassessment),usersmusthavesufficientcapacitiestounderstandthemechanicsandimplicationsofkeydecisionsregardingitsuse.Theseincludetheselectionofcriteria, definition of indicators, and relationships to other decision supportmethods. As such, capacity building remains essential. While relieving someanalyticaldemands,IPFneverthelessrequiressufficienttechnicalknowledgetoappropriatelyspecifyandcalibrate theadditivemodels,variables,andweights.Moreover, capacity‐building efforts to improve project appraisal and selectionshould also involve training on CBA, which is essential for larger projects, inparticular.

Further, to extoll the benefits of responsiveness inherent to the tool, theproposedmethodology should not be a one–off exercise. Rather, it should beutilizedasaprogressiveapproach,intendedto‘liveandgrow’withthecountry'sinfrastructure needs and policy objectives. As such, the prioritization programshould involve continuous refinement of the decision‐support tool, based on

penalized.TheMPIorsectoragencywillallocatefundingforFSstoprojectsinthehighestimportancegroupsandthenproceedthroughtheremaininggroupsuntiltheFSsfundingisexhausted.

20

informed deliberation regarding criteria selection and any pre‐decisions of apolicynature(Mandri‐Perrott,Marcelo,&Haddon,2015).

Fifth,governmentsmustbeawareofpotentialsequencingconflictsrelatedtothetiming of project selection processes in different ministries and line agencies.One strategy tomitigate this is for central government tooverseeandprovideguidance for systematizing project selection and coordinating sectorialprioritizationactivities.

Last,planningofficesanddecisionmakersmustbefamiliarizedwiththemulti‐criteriaapproachtobuildcredibilityofthedecisionsupporttoolitself,requiringconsultationandclearexplanationof the technical inputsandmodelstructure.While some fluency with established quantification methods like CBA isobserved,thereisacertaindegreeofriskaversionassociatedwithapplyingnewmethods. Building acceptance of the decision support tool itself is critical tolegitimizingtheanalysis(Mandri‐Perrott,Marcelo,&Haddon,2015).

Panama:TransportandWaterandSanitation

InPanama,theconflationofthecurrenteconomicoutlookandthreeinstitutionalsupports spurred the IPF pilot. GDP growth and economic buoyancy in 2014motivated an ambitious public investment program, accompanied by a highnumber of infrastructure project proposals to the Ministry of Economics andFinance. Nevertheless, the proposed set of projects exceeded the availablefundingspaceandallowabledeficitceiling,demandingselectionofsomeprojectsand postponement of others. The application of a prioritization methodologywas endorsed in the Government Strategic Plan 2015‐2019 and a draftamendment to the 2008 Social Fiscal Responsibility Law, which stated that asystemofprioritizationstrategieswasneededforinfrastructuredevelopmentinthefuture.Similarly,thepublicinvestmentlawcontainedimplementationnotesto employ a prioritization strategy tied to the five‐year investment plan 2015.The 2015 World Bank Country Partnership Framework (CPF) also called forapplication of a prioritization tool. These factors confirmed demand forprioritizationofinvestmentsininfrastructureplanning.

TheIPFwasappliedtoaselectionof35proposedprojectsinwatersupplyandsanitation and 19 in transport. These projects were identified in consultationwiththeMinistryofEconomicsandFinance.Thepilotofferedakeyopportunityto replicate and refine the existing framework, in that it entailed decisionanalysis basedon limited financial‐economic data, particularly for a portionofthe water projects. In this way, it replicated a common input problem forinfrastructuredecision‐making,namelyrestrictionsondata.

Planners from several agencies, in consultationwith theMinistry of EconomyandFinanceandateamfromtheWorldBank,agreedonasetofcomponentSEIand FEI variables. The SEI variables initially selected included the number ofbeneficiaries (BEN), direct jobs created during implementation (EMP), thepopulation of poor serviced by the project (POOR), social and environmentalrisks(SER),andthecarbonfootprint(CO2).Thefinalanalysisusedonlythefirstthree,duetodataproblemsandlackofspecificityintheriskvariable.17

17. SeeMarcelo,Mandri‐Perrott,&House,2015foradescriptionofthesechallenges.

21

The FEI was originally expected to include the internal rate of return (IRR)and/oreconomicrateofreturn(ERR)ofprojects,dependingondataavailability.However,giventhatmanyoftheseinvestmentswereproposedforprojectswithno direct monetary benefits and largely indirect economic effects (i.e. mainlyprojectswithalargepublicgoodcomponent18),thecalculationofeconomicIRRwouldhaveproducedunrealistic or incalculable results.Thealternativewouldhavebeentoaccountforall indirectpositiveeffectsandestimatebenefits.Thiswould have required data on monetized benefits for such effects, which wasunavailable.For thesereasons,benefit‐cost ratios (BCRs)wereused.TheBCRsalso allowed for the analysts to control for project size and avoid penalizingprojectswithhighercostsbutpotentiallyhigherbenefits.19

The results from the transport project SEI and FEI rankings are presented inFigure7.

Figure7.SEIandFEI,TransportProjects,PanamaPilot2015

Figure8showstheresultsofplottingeachproject’sSEIandFEIscores.Figure8.SEIandFEIplotforselectedtransportprojects,Panama2015

18. Anexampleofthissortofprojectwouldbeinvestmentsinwastewatertreatmentfacilities,wheretariffs

areofteninsufficienttocoverthefullcostsoftheassetsanditssustainedoperationandmaintenance.19. MoredetailsonthecalculationofcompositeindicesandresultsofthepilotinPanamaareavailablein

Marcelo,Mandri‐Perrott,&House,2015.

0

20

40

60

80

100

Proj

ect 3

Proj

ect 6

Proj

ect 2

Proj

ect 8

Proj

ect 9

Proj

ect 7

Proj

ect 5

Proj

ect 1

7Pr

ojec

t 19

Proj

ect 1

0Pr

ojec

t 13

Proj

ect 1

1Pr

ojec

t 18

Proj

ect 1

6Pr

ojec

t 4Pr

ojec

t 14

Proj

ect 1

Proj

ect 1

5Pr

ojec

t 12

Social and Environmental Indicator (SEI)

0

20

40

60

80

100

Proj

ect 4

Proj

ect 1

7Pr

ojec

t 5Pr

ojec

t 6Pr

ojec

t 18

Proj

ect 1

Proj

ect 7

Proj

ect 8

Proj

ect 2

Proj

ect 1

9Pr

ojec

t 13

Proj

ect 9

Proj

ect 1

6Pr

ojec

t 11

Proj

ect 3

Proj

ect 1

2Pr

ojec

t 14

Proj

ect 1

0Pr

ojec

t 15

Financial and Economic Indicator (FEI )

P1

P2P3

P4

P5

P6

P7P8

P9

P10

P11

P12

P13

P14

P15

P16

P17

P18

P19

0

10

20

30

40

50

60

70

80

90

100

0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 80.0 90.0 100.0

SEI

FEI

22

The pilot offered a number of key lessons and highlighted areas of futuredevelopment. The first andmost significant lessonwas that composite indiceswerefarmoresensitivetoindicatorvaluesthanweights.ThissuggeststhatPCAcan be a useful way to weight variables if time and objectivity are importantfactorsinselection.AsensitivityanalysiswasperformedtocomparePCAindicesagainstcompositeindicesusingsubjectivelyestablishedweights.Twosubjectiveweightingschemes(equalweightingandhypotheticalpolicy‐determined)weretested to calculate alternative SEI composite indices. Figure 9 shows that thecategorizationofprojectschangedonlyminimallywhenusingpolicy‐determinedorequalweights(Marcelo,Mandri‐Perrott,&House,2015).Inpractice,theuseof subjective weighting can give rise to a number of problems ranging frompurelytechnical(i.e.,rankreversalinAHP)topolitical(i.e.,lackoftransparencyand discretion in selection of infrastructure‐projects). On the other hand,subjective weighting is more intuitive and directly responsive to policypreferences.

Figure9.ComparisonofTransportSEIandFEIscores,PCAandequalweighting

Thesecondlessonisthatspecialconsiderationshouldbegiventotheselectionand definition of metrics to deal with regressive biases. This applies to anycomparative approach. As in Vietnam, Panama showed a potential for aninherent bias towards infrastructure projects in wealthier or urban regions,simply due to their generally better ‘performance’ on component indicators. Ifdevelopment plans were aimed at improving rural areas, however, this couldcreate adverse results. This problem can be overcome by careful indicatorspecificationortheinclusionofadditionalindicators.

The third lesson relates to the appropriate use of financial and economicindicators in conditions of low information. We recognize the value ofincorporating CBA elements on the financial and economic side of the IPF. Tointegrate these elements to the framework when information is extremelylimited,additionalcriteriaisrequired.Forexample,ifonlyprojectcostisknown,additional variables must be considered to build the FEI. This practicalconstraint highlights the importance of improving information systems at theprojectleveltoeffectivelyimplementanykindofprioritizationtool.

P1

P2P3

P4

P5

P6

P7 P8 P9

P10

P11

P12

P13

P14

P15

P16

P17

P18

P19

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100

SEI

FEI

PCA weighted SEI

P1

P2P3

P4

P5

P6

P7 P8P9

P10

P11

P12

P13

P14

P15

P16

P17

P18

P19

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70 80 90 100

SEI

FEI

Equal weighted SEI

23

The fourth lesson is that the IPF can take into account both efficiency andefficacy considerations. This is shaped by the construction of variables. Forexample,onecouldusetheabsolutenumberofbeneficiariesasaninputtotheSEItoconsiderpolicyeffectivenesswhenserviceexpansionisapriority.Ontheotherhand,‘beneficiariesperdollarspent’maybemoreappropriateifthegoalisfiscalefficiency.InthecaseofPanama,wheredevelopmentofwaterservicesinruralareaswasan importantpolicygoal, thedecisionwasmadenot tocontrolindicatorsbyprojectsizeinordertoprotectagainstthepossibilityofprivilegingurban projects with greater economies of scale (Marcelo, Mandri‐Perrott, &House,2015).

The final lesson is that IPF can be used as an opportunity to strengthen dataweaknesses. The process of discussing relevant indicators by which to selectprojects is a catalyst for improving information levels tomove towards morecomplex economic analysis. In Panama, the discussions prompted by the IPFhaveresultedinimproveddataprojectdatacollection.TheIPF,asaprocess,canbe a valuable starting point for improving appraisal by inciting aspirationaldiscussions about the kinds of data that would be most helpful to decidingamongprojectswithinasector.

ImplementingIPF

ThesequenceofimplementingIPF,asdescribedinthetechnicaldescriptionandpilot applications above, is summarized in Figure 10. With respect to thissequence, the pilot in Panama also gave rise to an important discussion ofrequirements for implementing IPFand factorsofparticular interest in furtherpilotingtoimproveitsapplicability.Theseissuesarediscussedinthefollowingsection.Someareprocess‐orientedandrelatetooverallinfrastructureplanning;some are organizational and deal with issues of resources, authority, andcapacity;andsomearetechnical,relatingtotheinformationalinputsrequired.

Figure10.SequenceoftheIPF

ICriteria

Identification • Deliberation and consultation with decision

makers and experts

IISEI and FEI calculation

• SEI and FEI based on existing appraisal data• PCA when objectivity in criteria weighting is preferred• CBA elements incorporated when available / calculable• Sensitivity analysis

IIIVisual interface

• (SEI, FEI) coordinates• Quadrants defined by budget

constraint

IVSelection

• Based on informed political and technical debate

• Quadrant A projects high priority to invest

24

Process

Withrespect to infrastructureplanning, it is importantthat IPF followon fromcritical pre‐prioritization and selection steps. These are described well in theWorld Bank’s Public Investment Management framework and include theprovision of clear sectoral policy guidance based on policy analysis, pre‐screening,andatleastbasicprojectappraisalandreview(Rajarametal.,2014,p.22‐26).Policyguidanceisalsoimportantforpre‐screeningprojectsforinclusionin prioritization exercises; important guiding documents include publishedsectorstrategiesandmid‐termdevelopmentplans.

Moreover,thereareimportantdecisionstobemadeaboutwhenIPFismoreorless applicable as a prioritization approach than are other approaches, e.g.,comparisonbyNPVfollowingfromSCBA.Forone,IPFmaybeusedtostructuredecisions when full economic analysis for all projects in the proposed set isunfeasibleortoocostlyforgovernment.Thereisanimportantquestionofhowmuch government should spend on appraisal and prioritization, relative toinvestment.Theissueofsettingthresholdsforappraisalrequirementsshouldbefurtherexploredtocreatecountry‐specificguidanceforappraisalrequirements.

OncethedecisiontoemployIPFismade,therearesomeimportantfactorsthatshouldbemadeexplicitandtransparentlyreportedinadvanceofimplementingtheapproach.Thesettingsthatmustbedecidedinadvanceinclude:selectionofcriteria and specific definitions, the weighting methodology to be used forconstructingindices,thetypesofsensitivityanalysistobeemployed,andguidingprinciples for selecting projects fromamongstQuadrantsB andC.After IPF isperformed and projects are prosed for selection, decision‐makers should alsopublishtheresultsoftheIPFmappingandsensitivityandreportdeviationsfromprojectselectionguidelineswithsupportingjustification.

Organization

While projects may be proposed from different line agencies or subnationalgovernment units, prioritization should bemanaged at the same level of fundallocationorbyanadministrativeunitwithauthorityoverinvestmentdecisionsto ensure that analysis is effectively utilized. Alternatively, governments canestablishlegislativeorotheradministrativebackingtoprotectthecredibilityofprojectselectionandensurethatselectionisactuallybasedonappraisalandtheagreedmethodofprioritization.

Itisalsoimportantthatprioritizationbealignedwiththebudgetcycle.TheIPFcan be used to prioritize capital investment plans for annual or multi‐yearbudgets. It is important, however, that the budget period and envelope beexplicitly determined and aligned, and that proposed projects be checked toensurethatinvestmentswouldfeasiblyoccurduringthespecifiedbudgetperiod.

Lastly,IPFisdesignedforapplicationwithinonlyonesector.Morespecifically,itmaybeadvisabletofurthersubdivideanalysisatthesubsectorlevel,particularlywhencertainkindsofprojects(inthecaseofwaterandsanitation,forexample)are significantlydifferent in termsofperformanceaccording to certainprojectindicators.

While IPF is intended for comparison of projects within a sector, it can be ausefuldevice fordiscussingsectoralreallocation.Theprojectplotscanbeused

25

toexamineimplicationsofforegoinginvestmentsthatperformatthemarginforanother sector’s budget envelope. For example, foregoing a large transportprojectatwithfundinginQuadrantBorCcouldfreeupsufficientbudgetspacetofundmultipleadditionalsanitationprojects.

MinimumDataRequirementsandImprovingDataQuality

While IPF is designed to make best use of available information and otherelements of project appraisal, there is aminimum amount of data required tomake theapproachusable.These includeestimationofprojectcostsaswellasmeasurements of expected benefits and key social and environmental projectimpacts.Projectcostsandsomebenefitscanbereportedasasinglemonetizedmeasure, preferably through the calculation of net present values via financialcost‐benefit analysisorpartial SCBA.Variables thatmeasureexpectedbenefitsandkeysocialandenvironmentalimpacts,ontheotherhand,maybereportedindifferentunitsorevenqualitatively.

SinceIPFisintendedtopromotetheuseofimproveddataandsupporteffortstoimprove infrastructure decision‐making overall, it is important that IPF notdetract from ongoing efforts to build overall capacity to appraise, select, andprepare good projects. Because the approach is complementary and notcompetitive with SCBA, for example, capacity building for good publicinvestmentmanagementshouldincludetrainingontheexpandedsetoftoolsforprojectappraisalandselection.

ConclusionsandNextStepsforDevelopingtheIPF

This paper presents the IPF as a flexible multi‐criteria tool for infrastructureprioritization.Theframework’skeystrengthsareitsincorporationofkeypolicygoals,thesystematicframingofinfrastructuredecisions,andtheuseofavailableelementsofprojectappraisal.Thegraphicaldisplayofresultsinasingleplaneisoneofthemainfeaturesofthistool.Thiswayofpresentingprojectinformationis helpful to informing discussions about relative expected performance ofprojectsaccordingtofinancial‐economicandsocial‐environmentalfactorsandtoconsideringthepotentialeffectsofreallocatingfundsbetweensectorbudgets.

TheexperiencesinVietnamandPanamashowthepotentialoftheIPFtosupportinfrastructure decisions, particularly in environments characterized by limitedinstitutional and technical capacity and data restrictions. It is important,however,toreaffirmthattheapproachisnotintendedtoreplacebestpracticesinprojectselection,suchasSCBA,orobviateeffortstoimproveappraisalinthepre‐selection stages of public investmentmanagement. Rather, the IPF can beused as a catalyst to identify needed information in order to progress towardmoresophisticatedappraisalmethodsandselectionframeworks.

LookingForward

WithrespecttotheIPF’sgeneralapplicability,wehaveidentifiedasetofcriticalissues for future deliberation to further improve the IPF. These include thefollowing:

‐ SectorialRebalancing:Howcandecision‐makersbestusetheIPFtoexplorepossibilitiesof rebalancing sectorial budgets?Where goodPPPcandidatesare identifiedwithin the high ormedium‐priority project sets, the budget

26

constraintmaybeextendedinthatparticularsector.Alternatively,fundsforprojects in quadrants B and C may be reallocated to other sectors withhighershortfalls,orhigh‐SEIscoredprojectswithinsufficientfunding.

‐ Participation and consultation:Which stakeholders should be included intheapplicationofIPF?Whataretheappropriateinstitutionalarrangementsandcheckstoensureaccountabilityandtransparency?

‐ Implementation:What institutional arrangements are required to supportimplementation of IPF and facilitate improved data collection? Given theflexibilityincriteriaselection,whatchecksarerequiredtoensurethetoolisappliedprofessionally,withsufficienttechnicaloversight?

‐ Sequencing / application to other project appraisal and selectionapproaches:Howcan IPFbeused sequentially or concurrentlywithothertools? What measures are needed to ensure that IPF is used to promotemorerigorouseconomicprojectappraisal?

Additionally,theIPFmaybeextendedinthefuturetoestablishlinkagestootherfacetsofinfrastructuredevelopment.Proposedextensionsincludethefollowing:

‐ PrivateParticipation:HowcantheIPFbeusedtoassistwith identificationof public‐private partnerships or other private financing opportunities?HowwillIPFresultschangewithvariablebudgetconstraintsresultingfromprivateparticipation?

‐ Integrated Planning: How can complementary projects be treated insimulation?

‐ Multi‐year Budgeting: How can IPF best deal with multi‐year budgetprocessesanddifferencesinsectorialbudgetingschedulesandprocesses?

Lastly,thereisaneedtodealwithprioritization,notonlyasaquestionofwhattodo,butalsoaquestionofwhen to invest.TheIPFmaybeusedtorulesomeprojects out altogether, but may also be extended to assess of the relativeimmediacy of proposed projects and timing of investments over long impacthorizons.IntheirassessmentoftheinfrastructureneedsinSouthAsia,Andresetal.suggestthatprioritizationcriteria“mustbeabletoanswersquestionsaboutshort‐term needs versus longer‐term development needs, especially indevelopingcountries…Givensubstantial lock‐insassociatedwithinfrastructureinvestments,shouldacountrycontinueattemptingtofillcurrentgapsordirectinvestments to infrastructures that are likely large bottlenecks in themediumterm?”(Andres,Biller,&Dappe,2014).

WhiletheIPF,asaflexibleframework,issufficientlymalleabletoaddresstheseextendedneeds,theanswerstotheseremainingquestionswillonlyberealizedthroughcontinuedpiloting.Refining this framework isaworthwhilepursuit toensure its appropriate application and usefulness to supporting infrastructuredecision‐makingwithaneyeto improveboththeefficient investmentofpublicfundsandthegoalsofsustainabledevelopment.

27

References

Andres, L., Biller, D., & Dappe,M. H. (2014). Infrastructure Gap in South Asia:Infrastructure Needs, Prioritization, and Financing. World Bank PolicyResearchWorkingPaper(7032).

Araral, E., Fritzen, S., Howlett, M., Ramesh, M., & Wu, X. (2012). Routledgehandbookofpublicpolicy.Routledge.

Bardach, E., &Patashnik, E.M. (2015).Apracticalguideforpolicyanalysis:Theeightfoldpathtomoreeffectiveproblemsolving.CQPress.

Barfod,M.B., Salling,K.B.,&Leleur, S. (2011).Compositedecision supportbycombining cost‐benefit and multi‐criteria decision analysis. DecisionSupportSystems,51(1),167‐175.

Beinat, E.,&Nijkamp,P. (1998).Multicriteriaanalysisforland‐usemanagement(Vol.9):SpringerScience&BusinessMedia.

Brown,H.(2014).Nextgenerationinfrastructure:Springer.

Cantarelli, C. Flyvbjerg, B., Molin, E., & Van Wee, B. (2010). Cost overruns inlarge‐scaletransportationinfrastructureprojects:explanationsandtheirtheoretical embeddedness. European JournalofTransport InfrastructureResearch,10(1),5‐18.

CDIA.(2010).CityInfrastructureInvestmentProgrammingPrioritisationToolkit:User Manual: From Wish List to Short List: Prioritising UrbanInfrastructureProjectsforLocalDevelopment.Manila,Philippines:CitiesDevelopmentInitiativeforAsia.

Dabla‐Norris, E., Brumby, J., Kyobe, A., Mills, Z., & Papageorgiou, C. (2011).Investing inPublic Investment:An indexofpublic investmentefficiency.IMFWorkingPaper11/37.Washington:InternationalMonetaryFund.

DCLG. (2009). Multi‐criteria Analysis: A Manual (pp. 165). London: LondonDepartmentforCommunitiesandLocalGovernance.

De Montis, A., De Toro, P., Droste‐Franke, B., Omann, I., & Stagl, S. (2004).Assessing the quality of different MCDA methods. Alternatives forenvironmentalvaluation,99‐184.

Dunn,W.N.(2015).Publicpolicyanalysis.Routledge,2015.

Freudenberg,M.(2003).Compositeindicatorsofcountryperformance.

Hajer, M., & Wagenaar, H. (2003). Deliberative policy analysis: understandinggovernanceinthenetworksociety.CambridgeUniversityPress.

Head, B. W. (2010). Reconsidering evidence‐based policy: Key issues andchallenges.PolicyandSociety,29(2),77‐94.

HMTreasury.(2014).UKNationalInfrastructurePlan2014.

Jacobs,D. (2008).AReviewofCapitalBudgetingPractices. IMFWorkingPaper08/160.Washington:InternationalMonetaryFund.

Jolliffe,I.(2002).Principalcomponentanalysis.JohnWiley&Sons,Ltd.

28

Kabir,G.,Sadiq,R.,&Tesfamariam,S.(2014).Areviewofmulti‐criteriadecision‐making methods for infrastructure management. Structure andInfrastructureEngineering,10(9),1176‐1210.

Khisty, C. J. (1996). Operationalizing concepts of equity for public projectinvestments. Transportation Research Record: Journal of theTransportationResearchBoard,1559(1),94‐99.

Layard,R.,&Glaister,S.(1994).Cost‐benefitanalysis.CambridgeUniversityPress.

Manaugh, K. (2013). Incorporating issues of social justice and equity intotransportationplanningandpolicy.McGillUniversity.

Mandri‐Perrott,C.,Marcelo,D.,&Haddon,J.(2014).AMethodologytoPrioritizeandSelectInfrastructureInvestments.ReporttotheVietnamMinistryofPlanningandInvestment.TheWorldBank.

Marcelo, D., Mandri‐Perrott, C., & House, S. (2015). Prioritizing InfrastructureInvestments in Panama: Pilot Application of the World BankInfrastructurePrioritizationFramework.Report to thePanamaMinistryofEconomyandFinance.TheWorldBank.

Martens,K. (2009).EquityConcernsandCost‐BenefitAnalysis:OpeningtheBlackBox.PaperpresentedattheTransportationResearchBoard88thAnnualMeeting.

Nardo, M., Saisana, M., Saltelli, A., Tarantola, S., Hoffman, A., & Giovannini, E.(2005).Handbookonconstructingcompositeindicators.

Nutley,S.M.,Davies,H.T.,&Smith,P.C. (2000).Whatworks?:Evidence‐basedpolicyandpracticeinpublicservices.MITPress.

OECD (2014). Recommendation of the Council on Effective Public InvestmentAcrossLevelsofGovernment.Adopted12March2014.

Perreault Jr, W. D., & Young, F. W. (1980). Alternating least squares optimalscaling: Analysis of nonmetric data in marketing research. Journal ofMarketingResearch,1‐13.

Petrie,M.(2010).Promotingpublicinvestmentefficiency:Asynthesisofcountryexperiences. Paper presented at World Bank Preparatory Workshop,PromotingPublicInvestmentEfficiency,GlobalLessonsandResourcesofStrengtheningWorldBankSupportforClientCountries.

Poister, T. (1978). Public program analysis:Applied researchmethods (p. 42).Baltimore,MD:UniversityParkPress.

Rajaram,A.,Tuan,M.,Biletska,N.,&Brumby,J.(2010).ADiagnosticFrameworkfor Assessing Public Investment Management. World Bank PolicyResearchWorkingPaper5397.TheWorldBank,August2010.

Rogers,M.,&Duffy,A.(2012).Engineeringprojectappraisal:JohnWiley&Sons.

Ruiz‐Nuñez, F., & Wei, Z. (2015). Infrastructure Investment Demands inEmerging Markets and Developing Economies. World Bank PolicyResearchWorkingPaper,7414.

29

Saaty, T. L. (2004). Decision making—the analytic hierarchy and networkprocesses(AHP/ANP). Journalofsystemsscienceandsystemsengineering,13(1),1‐35.

Salling, K. B., Leleur, S., & Jensen, A. (2007). Modelling decision support anduncertainty for large transport infrastructure projects: The CLG‐DSSmodelof theØresundFixedLink.DecisionSupportSystems,43(4),1539‐1547.

Thomopoulos, N., Grant‐Muller, S., & Tight, M. (2009). Incorporating equityconsiderations in transport infrastructure evaluation: Current practiceand a proposed methodology. Evaluationandprogramplanning, 32(4),351‐359.

Tsamboulas, D. (2007). A tool for prioritizing multinational transportinfrastructureinvestments.TransportPolicy,14(1),11‐26.

VanDelft,A.,&Nijkamp,P. (1977).Multi‐criteriaanalysisandregionaldecision‐making(Vol.8):SpringerScience&BusinessMedia.

Wang,Y.M.,&Elhag,T.M.(2006).AnapproachtoavoidingrankreversalinAHP.DecisionSupportSystems,42(3),1474‐1480.

Weimer,D.,&Vining,A.(2015).Policyanalysis:Conceptsandpractice.Routledge.

Zerbe,R.O.,&Bellas,A.S.(2006).Aprimerforbenefit‐costanalysis:EdwardElgarPublishing.

30

Annex1.DemandsofInfrastructureProjectPrioritization

Attention to alternative approaches to systematic prioritization responds todemands for evidence, comprehensiveness, value, efficiencyand legitimacy.Animportantunderpinning forprioritization is the consideration ofevidence. Therise of evidence‐based policy analysis (EPBA) can be traced to the UKgovernment of the 1990s, which rigorously pursued approaches to policyanalysis in cross‐departmental teams working on complex issues. EBPA isbelievedtobeimportanttoanswerquestionslike:Whatoptionswill‘deliverthegoods’? How can programs provide greater ‘value for money’? and ‘How canprogrammanagersachieve‘outcomes’?(Nutley,Davies,andSmith,2000).

Expandingbeyondpurelytechnicalresearchandanalysis(i.e.,‘science’),alineofthoughtrecentlygainingtractionsuggests that twoothertypesofevidencearerelevanttomodernpolicymaking:practice(professionalprogrammanagementexperience) and political judgment (Head, 2010). Scientific knowledge isgarneredfromsystematicanalysisofcausalrelationshipsthatexplainconditionsand trends.Practical implementationknowledge,on theotherhand, isderivedfrom the “practicalwisdom” of professionals in their communities of practice,whereaspoliticalknowledgerelatestomakingcontextualjudgmentsabout“thepossibleandthedesirable”(Head,2010,p.80).AnincreasedappreciationofthemultiplebasesofknowledgeisthenextiterationofEBPA.

Anappreciationof the ‘multiple lenses’ofevidencesuggests thatprioritizationmust also be sufficiently comprehensive. Comprehensiveness implies makingdecisions based on (a) consideration of a sufficiently extensive set of projects,and(b)asufficientlywidesetofcriteria.Whilecriterianeednotbeexhaustive,they should nevertheless account for key goals of infrastructure policy. Directattendance to themultiple goals embedded in an infrastructure strategy (e.g.,fiscal prudence, equity, sector‐specific gains, etc.) supports the use ofmultiplecriteria. Furthermore, comprehensiveness is a response to common criticismsthat infrastructure investment decision‐making is ad hoc, politically driven, orcharacterizedby‘cherry‐picking’,‘easywins’,or‘creaming’ratherthananalysis.

Decisions about infrastructure investment are also inherently integrated withconsiderations of effectiveness and value. The selection and structuring of aprioritizationmethodologyare intertwinedwithdecisionsabouthowtodefinepolicyeffectiveness,whichmayincludegoalsofeconomicgrowth,sectorialgoals,environmentalsustainability,orhumandevelopment.Thelogicofvalue,ontheotherhand,speakstocreatingpublicvalueattheleastcost.

Lastly, prioritization is undertaken to afford public legitimacy to decisions,particularly in actively democratic contextswhere public assentmattersmost.Legitimacy is typically founded on both inputs and outputs. Input legitimacyreferstotheprocesseswherebydecisionsaremadeandisamatterofdesign.Tobelegitimatewithrespecttoinput,theprocessofinfrastructureselectionshouldbe transparent, fair, and systematic. Output legitimacy, on the other hand, isdeterminedbyoutcomes,andisamatterofaninstitutionearningitsrelevancebasedonperformance.

While not the primary focus of this paper, a final parameter of interest isopportunity.Inthecaseofinfrastructureprioritization,thisreferstotheopening

31

of newopportunities for funding.As options for infrastructure finance expandbeyondpublicresourcesandtraditionalbanklending,moreattentionisfocusedonconditionsthatincreasethelikelihoodofinstitutionalinvestmentandprivatesectorparticipation.Theassemblageofwell‐plannedprojectpipelines,whichareprioritizedinalegitimateandtransparentfashioncanhelpreducepoliticalrisksandimproveprojectbankability.

Theprioritizationapproachagovernmentadoptsmustinevitablybalancethreeoft‐competingneeds:accuracy,feasibility,andsuitability.Accuracydemandsthatmethods be sufficiently precise to affordmeaningful comparison, but does notrequireextremeexactitude.Rather,itsuggeststhatthresholdsofcorrectnessarerequiredtoensurethatthelogicofevidenceisattained.

The second condition encompasses administrative practicality and politicalfeasibility. Practicality deals with the institutional capacity, the cost and timelimitations, and the information availability. An important facet of this is theprinciple ofparsimony, the use of aminimumamount of relevant information.Political feasibility, on the other hand, accepts that prioritization cannot be sodevoidoflatitudethatitisrenderedunresponsivetopoliticalfactors.Thereisabalancetobestruckbetweentechnicalobjectivityanddemocraticaccountability.

Lastly, theprincipleofsuitabilitydemandsthatcriteriaselectedbeappropriateto judge relative desirability, as determined by stakeholders and theirrepresentatives.Thesuitabilityofdecisioncriteriaisdependentonpolicygoalsandnormsofgovernance.

32

Annex2.Cost‐BenefitAnalysisandValueforMoney

Cost‐benefit analysis (CBA), is a useful method of project comparison wherecapacityanddataaresufficientforitsimplementationacrossproposedprojects.The practical development of CBA started in the 1930s in the United States,largely forpublic investmentplanningat the federal level, andhas remainedastapleofpolicyanalysissince(Zerbe&Bellas,2006).ThisislargelybecauseCBA,whilst complex with respect to inputs, is a straightforward concept, allowingcomparisonofprojectsbasedonasinglemetric–thatofmonetizedvalue.

CBA essentially totals all costs and benefits of a project over its lifetime anddiscounts future flows to calculate present values. The (discounted) presentvalues of costs and benefits are compared, either by use of net present value(ranking projects by highest NPV) or the benefit‐to‐cost ratio (BCR) (used toreflect efficient use of inputs for outputs). CBA can be applied in traditionalfinancialtermstoassessalternativeprojectsforanorganizationorfirm,butmayalsobeextendedforpublicexpenditureanalysisbyconsideringthefullsuiteof(monetary and non‐monetary) costs and benefits to society. In Social Cost‐Benefit Analysis (SCBA), prioritization is based on selecting projects thatmaximizethenetpresentvaluesforsocietyoverall.Reductiontoacommonunitofmeasurementallowseasycomparisonofprojects(Andresetal.,2014).

Theseassessments require thequantificationandmonetizationofpositiveandnegative effects (costs and benefits); therefore, extensive information aboutprojects and their projected impacts is required (Van Delft &Nijkamp, 1977).Since information inmany contexts is limited, however, and sincemany costsand benefits are difficult to monetize20, this can make SCBA a tall order,particularlywhengovernmentswithlimitedresourcesarepresentedwithlargesets of small and medium size projects for comparison. Practically speaking,relevant data for SCBA may be unavailable, too expensive, or too difficult tocollect or calculate on a regular basis, given a government’s financial andtechnicalresourcesforprojectpreparation.

WhileanumberoftechnicalissuesrelatedtoCBAmechanicsarecommonfocalpoints of academic debate,21these can be largely dealt with via establishedeconomic methods to yield relatively robust analyses. More important to thesystematicuseofCBAforinfrastructureprioritizationaretheresource,capacity,andtimelimitationsthatconstrainitsextensiveapplicationinmanycontexts.

20. For example, valuing lives saved, added convenience, or averted pollution requires making

assumptions. A common criticism of SCBA is that social and environmental costs are oftenunderestimated, relegating key social and environmental issues to positions of lesser importance,particularlyiftheirmonetizedimpactsarerelativelylowcomparedtoothereconomicconsiderations.

21. First,addressingintangiblefactorsandstrategicconcernsisdifficultwithCBA.Commonprocedurestoestablishmonetizedvalues for somenon‐marketed factors(e.g., statedpreferenceorhedonicpricingapproaches) are not necessarily applied to all non‐priced impacts (e.g., social cohesion). Theseproblemsexposeapotentialforoptimismbiasorcostunderestimation(Cantarelli,Flyvbjerg,Molin,&VanWee,2010;Thomopoulosetal.,2009),andrepresentinformationgapsthatcanbedetrimentaltoanalysis.Asecond issuerelates to the selectionofadiscount rate.Manyanalysesassumeastandardratetobeappliedtoacountryandsector,whichundoubtedlyalleviatestheburdenofdetermination.However, it is also known that slight alterations in rates of return can have a significant effect oncalculatedbenefit–costratiosandnetpresentvalues(Thomopoulosetal.,2009;VanDelft&Nijkamp,1977).

33

An additional set of considerations has prompted adjustments to SCBA or thecomplementingofSCBAwithotherapproaches.Forone,CBA‐basedassessmentsof societal value do not typically consider distributional effects or issues ofequityandsocialjustice,whichisakeyconcernwheninvestmentsareintendedtoclosedevelopmentgaps.Inotherwords,thegoalismaximizingsocietalvalue,butwithoutregardtotheparticular‘winners’and‘losers’ofalternativeprojects.Complementary social analysis or the use of multi‐criteria methods to extendCBAhavebeenusedtodealwiththeseaspects.

Other attempts are made to extract the particular costs and benefits that arelinked to key priority goals from the overall CBA in order to focus morespecifically on partial impacts of key policy interest. For example, Berechmanand Paaswell (2005) utilize a modified SCBA to assess transport projects,separatingouttheestimationsoftransportationbenefitsandcostsfromoveralleconomicdevelopmentbenefitsandcosts.Theapproachwasemployedtodealwith the observation that overall economic costs and benefitswere far higherthantransportation‐specificcostsandbenefits(e.g., timesaved).Thisrenderedthe latter completely insignificant to the overall assessment, despite theimportanceoftransportgoalstopolicymakers.

‘ValueforMoney’(VfM)referstoabroadrangeofapproachesused forprojectappraisal,manyofwhichwhollyorpartiallyemployCBA.VfManalysesmaybequalitative or quantitative, but generally compare projected project outcomeswith the resources employed to attain them. In this way, VfM is useful as anassessmentof the relativeefficiencyof alternativemeansof reaching thesameends. That said, VfM analyses are widely disparate with respect to the inputsused and level of quantification. Some are inclusive of full CBA or cost‐effectivenessanalysis22askey inputs,whereasothersmayqualitativelydiscussthedifferencesbetweenalternativesofthesamecostordifferencesinbasecostsforalternativesolutionstothesamepolicyproblem.

22. Costeffectivenessanalysisrelatestoevaluatingalternativemeansofattainingthesamegoals,butisless

applicabletodifferentiatingbetweentherelativeattractivenessofdifferenttypesofprojects.

34

Annex3.Multi‐criteriaanalysisforinfrastructureselection

Multi‐criteria approaches to infrastructure development and planning respondto concerns about over‐specialization, the need to reconcile multipleinfrastructure‐relatedpolicygoals,andpracticallimitationsoninformation.Thepast ten years havewitnessed an opening tomulti‐criteria analyses to rectifytheseconcerns,makedirectuseofpoliticalknowledgeandjudgment,andbringnon‐financialconcernstotheforefrontofdecision‐making.

On thepractical side, increaseduseofMCDAreflectsattempts toworkaroundtime,information,andcapacityrestrictions(DCLG,2009).Ifinformationisfullyavailable and the policy goal at hand ismaximization of societal benefit, thenoptimizingmodelsaremostuseful.23Butifinformationisincompleteormultiplepolicy goals are at stake, Beinat and Nijkamp suggest that use of a variety ofdecision criteria reflects compromise between multiple priorities, whilediscrepanciesbetweenoutcomesandgoalsare tradedoffbyuseofpreferenceweights (Beinat &Nijkamp, 1998).MCDA has been particularly salient for thetransportsector(Tsamboulas,2007),sinceprojectsoftenhaveahostofdifferentobjectivesandcomewithcomplexissuesarisingfromtheproject.

Intermsofintuitionandtransparency,additiveMCDAmodelsthatsumcriteriawithassignedweightsarefavorablesincetheyare“abletocopewithalmostanyproblem”(Tsamboulas,2007)andareeasytounderstand.WhilesomeadditiveMCDAsarecontingentonwhollyquantitativeandstatisticalinputs,others,suchas the Analytic Hierarchy Process (AHP), focus explicitly on expert valuejudgmenttoassignvaluestovariablesandcriteriaweights.

De Montis et al., extensively compare alternative decision methods forapplicationtosustainabledevelopment,contrastingtheapproachesacrosstheiroperational components, applicability in the user context, and applicability fortheproblemstructure (2004).They find that important considerations includethe ability for the approach to dealwith complex situations (criteria, differentscales and aspects, type of data, uncertainties); thepossibility to involvemorethan one decision‐maker (stakeholder participation, communication, andtransparency);andengagementofstakeholdersinordertoincreaseknowledgeand change opinions on problem structuring and alternative solutions (DeMontisetal.,2004).

Twocritical issues intheapplicationofMCDM,specifically foradditivemodels,are the selection of criteria by which alternatives will be assessed and theweighting of criteria. The simplest mode of weighting is equality, wherein allcriteria are equally considered. For example, the Cities Development InitiativeforAsia equallyweights (.20)project purpose, public response, environmentalimpact,socio‐economicimpact,andfeasibilityofimplementationinitstoolkitforassessing public projects (CDIA, 2010). An alternative method is negotiatedexpertguidance,whereinapanelofdecision‐makersdecidesweightsbasedonexperience and their basket of interests. A formalized, facilitated method ofexpert‐based criteria weighting is the Analytic Hierarchy Process (AHP),

23. For the principle of optimization to be the basis of a decision, however, itmust be assumed that all

measuresofperformancerelatedtotheobjectivesofaproposalcanbeexpressedinacommonscaleofmeasurement,asinCBA(Rogers&Duffy,2012).

35

developedbySaaty(2004).ThekeyinputforAHPisdecision‐makerresponsestoaseriesofpairwisecomparisonsofalternativeoptions.Responsesareusedtoderivecriteriaweightsandperformancescores.AHP isusedextensively in theRepublicofKoreaforinfrastructureplanningtosupplementSCBAanalysis.24

Through an extensive review of journal articles focused on MDCA andinfrastructure decision‐making between 1980 and 2012, Kabir, Sadiq, andTesfamariam find that the application of multi‐criteria approaches toinfrastructure selection and evaluation are on the rise. The number of MCDAinfrastructure management studies has risen from single digits in the 80s tohundreds towards the end of the 2000s (Kabir, Sadiq, & Tesfamariam, 2014).WhilstthesedonotnecessarilymeasureMCDAinuseforselection(andarefarfewer than studies based on CBA), the pattern indicates a sharp upswing ininterestinMCDAforinfrastructure.Ofthese300studies,however,onlyasubsetof approximately 40 expressly applied MCDA to choose amongst alternativestrategies for infrastructuremaintenance anddevelopment (e.g., site selection,project architectures alternatives, technology selection), butwithmore limitedscopesthansector‐wideprojectprivatization.Onlyfourteenattendedbroadlytoranking projects (as opposed to selecting amongst mutually exclusivealternatives),andonlytenrankedprojectsacrossasector.Thissuggeststhatthespace for developing sectorial infrastructure prioritization strategies remainswideopen.

Figure11.Numberofstudiespublishedrelated toMCDMin infrastructuremanagement,1980‐2012(Kabiretal.,2014)

Source:Authors’figure,drawnfrom(Kabiretal.,2014)data

24. AHPhasalsohadsomeweaknesses.Forexample,thereisamethodologicalproblemof“rankreversal”,

whichoccurswhenanadditionalnewoptionbecomesavailablethatresultsinthereversaloftheinitialalternatives’ranking.Therehavebeenattemptstoovercomethis,however,withvariousimprovementswhilstretainingtheunderlyingstrengthsoftheapproach(e.g.Wang&Elhag,2006).Theapproach issupported by well‐known adaptations with corresponding software applications, includingREMBRANDT,MACBETH,ExpertChoice,andorHIPRE.

0

20

40

60

80

100

1980‐1982

1983‐1985

1986‐1988

1989‐1991

1992‐1994

1995‐1997

1998‐2000

2001‐2003

2004‐2006

2007‐2009

2010‐2012

Others

Underground infrastructure

Buildings

Bridges

Transportation

Water supply and wastewater

Water resource management

Total

36

Annex4.Examplesofgloballyrecognizedindices

Area/Sector CompositeIndicator

Economy CompositeofLeadingIndicators(OECD)OECD International Regulation Database(OECD)EconomicSentimentIndicator(EC)InternalMarketIndex(EC)BusinessClimateIndicator(EC)

Environment Environmental Sustainability Index (World EconomicForum)WellbeingIndex(Prescott–Allen)SustainableDevelopmentIndex(UN)SyntheticEnvironmentalIndices(IslaM)Eco–Indicator99(PreConsultants)ConcernaboutEnvironmentalProblems(Parker)IndexofEnvironmentalFriendliness(Puolamaa)EnvironmentalPolicyPerformanceIndex(Adriaanse)

Globalization

GlobalCompetitivenessReport(WorldEconomicForum)TransnationalityIndex(UNCTAD)GlobalizationIndex(ATKearny)GlobalizationIndex(WorldMarketsResearchCentre)

Society HumanDevelopmentIndex(UN)CorruptionPerceptionsIndex(TransparencyInternational)OverallHealthAttainment(WHO)NationalHealthcareSystemsPerformance(King’sFund)RelativeIntensityofRegionalProblems

InnovationandTechnology

SummaryInnovationIndex(EC)NetworkedReadinessIndex(CiD)NationalInnovationCapacityIndex(PorterandStern)InvestmentinKnowledge–BasedEconomy(EC)PerformanceinKnowledge–BasedEconomy(EC)TechnologyAchievementIndex(UN)GeneralIndicatorofScienceandTechnology(NISTEP)SuccessofSoftwareProcessImprovement(Emam)

37

Annex5.SelectApproachestoPrioritization

UnitedKingdom

The UK National Investment Plan, managed by the Treasury’sinfrastructureunit, specifies an InfrastructureTop40 list of projectsmarked for priority government support and investment. Theseprojectsaregroupedbysector,butnotlistedinorderof importance.Projectsarechosenbythefollowingcriteria:

‐ Strategic importance (SI): significant contribution towardsanobjective;

‐ Capitalvalue(CV):significantcapitalvalue;‐ Regionalpriority(RP):highstrategicimportanceorcapital

valueinaregion;‐ Demonstrator (D): innovative or novel and could improve

futuredelivery;‐ Unlocking investment (UI): enables significant private

sectorinvestment(HMTreasury,2014).

Australia

Infrastructure Australia, a federal statutory board established undertheDepartmentInfrastructureandTransport,istaskedwithplanningand coordination cross‐state road and public transport projects. Inorder to prioritize proposed projects, the agency applies a two‐stateprocessofproject“profiling”and“appraisal.”Profiling,asafirstfilter,qualitatively assesses the compatibility of proposed initiatives tostrategicinfrastructurepriorities(i.e.,keyissuesandproblems)alonga scale of “highly beneficial” to “highly detrimental” with respect tostated policy goals. Thereafter, CBA is employed as the primary toolforprojectappraisal, includingestimatesofWiderEconomicBenefits(WEBs),suchasthoserelatedtoagglomeration.AdviceoncalculatingWEB is based on the UK government’s Transport Analysis Guidance(2014). Following CBA, the process requires that assessorsqualitatively discuss benefits and costs that generally cannot bemonetized (e.g., visual / landscape, social cohesion, heritage orcultural impacts) and thereafter classify each non‐monetized itemalong a spectrum from “highly beneficial” to “highly detrimental.”These two inputs are used to inform selection, which is based onexpertreviewandconsensusofapanelofelevenmembers.

Australia, NewSouthWales

New South Wales has developed a Major Projects AssuranceFramework inclusive of an additive multi‐criteria model. Theframework assesses proposed projects at several stages of projectplanning and prioritizes projects according to assessed performancealong two dimensions. Performance with respect to strategicobjectives ismeasuredbyalignmentwithNSW’s investment themes,value formoney, theproject’s ability to afford citizens “a better life”(by reducing cost of living and improving livability), and economicefficiency. Performance with respect to the ‘Infrastructure NSWProject Assurance’ objective is based on sufficiency of the analysis,cost‐benefit analysis, professional assessments of the suitability ofproject management, and risk assessment. CBA is augmented byprofessionalreviewandqualitativeinputs.

Qualitative assessments are numerically scored on a scale from 3(stronglypositive)to‐3(stronglynegative)andaddedusingasystem

38

of weights decided by a panel of professionals within InfrastructureNSW. Similar to the proposed IPF, projects are plotted onto a two‐dimensional plane, with axes defined by the Strategic Objective andProject Assurance Objective scores. Projects are classified as short‐,medium‐,andlong‐term,dependingontheircollectivescores.

InfrastructureNSWconceptualprojectmapping

Source:GovernmentofNewSouthWales(2014)

Korea,Rep. TheRepublicofKoreaemployscost‐benefitanalysis,supplementedbymulti‐criteria decision methods to prioritize a large number ofprojects across sectors. Using the AHP structured expert pairwisetechnique, experts decide the weights of decision criteria, includingSCBA. AHP has also been used to rank projects sub‐sectorally(primarilyintransport)intheUS,Indonesia,China,Turkey,India,andPalestine, but is not (to our knowledge) used as a nationalprioritizationframeworkoutsideKorea.

Indonesia During2014‐2015,Indonesia’sCommitteeforAccelerationofPriorityInfrastructureDelivery(KPPIP)employedathree‐levelinfrastructureprioritization approach, includingmulti‐criteria analysis. Following ascreening for basic project requirements, an additive multi‐criteriamodelwastoidentify22priorityinfrastructureprojectsfromamongstthousandsofproposedprojects.Theindicatorsforprojectscoringandranking (with associated additiveweights) included project purpose(25%); feasibility of implementation (30%); socio‐economic impact(30%); and environmental impact (15%). The scoring and rankingoutcomeswereusedasabasisof“committeediscussion”thatresultedintheshortlistingof22projects.

CitiesDevelopmentInitiative(internationalpartnership)

TheCitiesDevelopmentInitiativepublisheda2010CityInfrastructureInvestment and Prioritisation Toolkit that utilizes a multi‐criteriadecision model to consider project purpose; public response;environmental impact; socio‐economic impact; and feasibility ofimplementation. The prioritization approach was designed toprioritize funding agency support for projects proposed by localmunicipalities.

39

Annex6.PrincipalComponentAnalysis

PrincipalComponentAnalysis(PCA)isaninformationreductionprocedurethatseeksredundancieswithinasetofvariables(Joliffe,2002).Theseredundanciescan be expressed as linear combinations or ‘principal components’ of thevariablescomprisingtheset.Eachprincipalcomponentisaweightedaverageofthe original indicators or component variables. The coefficients, or weights,associatedwithvariables ineachprincipalcomponentare thosethatmaximizethevarianceofeach.

Thefirstprincipalcomponentcorrespondstothelinearcombinationofvariablesthatretainsthemaximuminformationoftheoriginaldataset.Thenotationforthefirstprincipalcomponent is:

(1) ⋯

Where denotes each observation and denotes the weight for thevariable .

In the context of the IPF, the first principal components of the social‐environmental and financial‐economic variable sets become the compositeindices.The coefficients of each ‘firstprincipal component’ are takenas theweightsassociatedwitheachvariable .

StatisticalsoftwaresuchasSPSSandSASincluderoutinestoperformPCA.