5. Scenario Analysis: A Tool for Task Managers · 5. Scenario Analysis: A Tool for Task Managers...

62

5. Scenario Analysis: A Toolfor Task ManagersJonathan N. Maack

Scenario analysis has been used by the privatesector for the last 25 years to manage risk anddevelop robust strategic plans in the face of anuncertain future. Its success in helping firmsmanage large capital investments and changecorporate strategy has made it a standard toolof medium- to long-term strategic planning.Scenarios have helped public sector agenciesplan for population growth and regionaldevelopment, state transportation investments,and the distribution of landfills. In the develop-ing world, scenarios have been used to high-light the opportunities, risks, and trade-offs innational policy debates.

Uses

Scenario analysis is different from the high,medium, and low cases for lending consideredas part of Country Assistance Strategy (CAS).High, medium, and low scenarios look atdifferent rates of progress along a path that isbased on a single set of projections. Scenarioanalysis focuses on the areas of greatest uncer-tainty for a country or an operation, systemati-cally develops several plausible alternativefuture environments in which the operationmight be implemented, and determines howthey would affect its success. This structuredapproach to thinking about the future could helpWorld Bank country teams make strategic choicesabout where and how to direct lending over themedium term in several ways:

Managing risk. Scenarios that look at futurepaths for a country help decisionmakersmanage risk and develop concrete contingencyplans and exit strategies. Describing how andwhy possible futures might occur enablesdecisionmakers to reflect on how political,social, and economic changes affect the opera-tion and to plan accordingly.

Building consensus for change. Scenarios are alogically rigorous, transparent means to givestakeholder representatives a role in planning.Because a number of possible futures arecreated, many perspectives can be included,and the discussion does not revolve around theadvocacy of fixed positions. All participantstake part in formulating the core elements of allscenarios, and later break into smaller groupsto write each of the four or five scenarios. Thisprocess helps decisionmakers share theirthinking about uncertainty and risk, developmutually understood contingency plans, anddefuse blame-casting when forces outside theircontrol lead to a change in strategic direction.

Augment understanding about the future.Scenarios are designed to bring up issues thatotherwise would not be considered by expos-ing the underlying forces in a sector or region.They are most effective when dealing with bigissues and strategic directions, rather thantactical decisions, and should not be used forshort-term planning.

63

Scenario Analysis: A Tool for Task Managers

Monitoring progress and scanning changes inthe environment. Scenario analysis can helpestablish indicators that create a framework tomonitor the execution of a strategy. Decision-makers then “scan” the external environmentfor the agreed indicators that the country orregion is beginning to move toward a differentscenario (or to a different stage along a scenariopath). The focus on key drivers of changemakes explicit the assumptions underlyinglending and speeds mobilization of resourcesin response to changes in the external environ-ment.

Scenario analysis is a tool with a strong trackrecord and significant potential, but it needs tobe adapted and piloted before it can be morewidely used in the Bank. This chapter describesa scenario process designed to be the corner-stone of a strategic plan. It also suggests waysto use the key concepts of scenario thinking inshorter exercises designed to help decision-makers minimize risk, address key uncertain-ties, and more effectively share their thinkingwith stakeholders about a lending programand the country’s future.

Process

Scenario analysis follows a systematic processto create a set of four to five plausible andvividly contrasting narratives that describepossible evolutions of key areas of uncertainty(figure 1). These narratives, known as sce-narios, examine the social, political, economic,

and technological forces that will impact on aproject or strategy. Scenarios are arrived at by ateam composed of key decisionmakers, experts,and stakeholder representatives during two orthree one-day workshops held over a period ofweeks or months. Because they focus on areasof uncertainty and the potential for unexpectedfuture discontinuities, scenarios provide aperspective not captured through projectionsbased on past data.

Scenario analysis involves constructing ordeveloping scenarios (steps 1-4 below), andintegrating the content of scenarios into decision-making (steps 5-8 below). During the exerciseitself, it is important that both of these elementsreceive equal attention and that mutuallyacceptable triggers and measures of success beagreed as outcomes of the process.1

1. Define a focal issue or decision. Scenarios arebest suited to looking at the future through thelens of a specific issue, such as the likelyoutcome of a national election leading to atransfer of power or a currency devaluation.Without this grounding, there is a danger thatthey will be too general.

2. Identify driving forces. These are the social,economic, environmental, political, and techno-logical factors that are most relevant to thefocal issue. They should be prioritized by thescenario team according to their level of pre-dictability and importance in affecting thedesired outcome.

Figure 1. Diagram of the Scenario Process

Choose scenario group Preliminary interviews

Preliminary research

First scenario 4-5 draft scenarios Verify scenario Write full-length workshop(s) assumptions scenarios

4-5 final scenarios

Review and revise scenarios Implement operationand monitor progress

Communicatescenarios

Indicators forexternal events

in each scenario

Strategicimplications ofeach scenario

Final scenarioworkshop(s)

Legend :Core scenario teamFull scenario team

64

Jonathan N. Maack

3. Write scenario plots. These are the storiesthat explain how driving forces interact andwhat effects they have on the operation orstrategic direction being discussed.

4. Flesh out scenarios. Good scenarios combinea solid understanding of relevant presenttrends with a clear focus on the outcomessought by decisionmakers. They should incor-porate lessons revealed by analysis of quantita-tive and qualitative data relevant to theassumptions underlying each scenario.

5. Look at implications. After scenarios havebeen fully developed, decisionmakers shouldstudy their implications for the outcomes beingsought by the operation.

6. Choose “leading indicators.” These indica-tors should help decisionmakers monitorchanges in the external environment as well asdevelopments in the project.

7. Disseminate scenarios. Once scenarios have beenbuilt and refined, they should be written in succinct,easy-to-read language and disseminated withinimplementing organizations and to the public.

8. Integrate scenario outcomes in daily proce-dures. Change the incentive system in affectedagencies and areas to ensure concerted move-ment toward the strategic goals that have been

formed on the basis of the scenarios. Use ormodify existing systems to monitor progresstoward operational goals as well as changes inthe external environment.

Scenarios in Practice

The two primary situations in which scenarioanalysis has been used by the public andprivate sectors are:

1. When considering a significant capitalinvestment decision or sectoral strategy

2. When developing or reformulating na-tional or corporate strategy.

Scenario Analysis for Capital Investments

Scenario analysis is extensively used in theprivate sector by companies that make largeinvestments in productive capital or researchand development. Utilities, hospitals, oilcompanies, and pharmaceutical companies allinvest significant amounts of money in re-search, development, and site construction.They need to make all possible efforts to ensurethat their investments will remain viable overthe long term. Scenario analysis has helpedthem examine the political and regulatoryenvironments, the likely attitudes of consum-ers, and other relevant factors to attempt toensure that their investments continue to beprofitable as conditions change (box 1).

Scenarios to assess the riskiness of an investmentstrategy need not be simplistic. The process ofidentifying the real focal issue behind an investmentis often quite revealing. For example, Shell OilCompany has used scenario analysis since the1960s. When Shell began looking into investing innew oil fields in the late 1970s, it asked its scenarioteam to evaluate the investment in terms of forcesthat would affect the price of oil in Western Europe.

The team came back with several scenarios. Onecited the risk of the collapse of the Organization ofPetroleum-Exporting Countries (OPEC) due tointernal tensions and external pressure. Anotherlooked at the demographics and political situation inthe Soviet Union and noted that a relaxing oftensions with the West could lead to large amountsof oil from Central Asia being exported to Europe.

Box 1. Scenario Analysis for Capital Investment: Shell Oil Company

Another looked at a future in which the pressure ofhigh prices and environmental organizations led totechnological breakthroughs that reduced thedependence of Western Europe on oil and gaso-line.

All of these scenarios were plausible, and,indeed, all did come to pass to a degree. Shellplanned its investments to take advantage ofpotential technological breakthroughs and to growthrough a flexible series of small expansions. WhenOPEC was no longer able to control enough of theoil supply to keep prices high, Shell’s conservativeposition allowed it to rise from seventh in the industryto third. Its small investments remained profitablewhile competitors lost significant amounts of moneyas riskier oil fields tapped during a period of highprices failed to perform.

65


In the public sector, scenarios have been usedto guide large capital investments in transpor-tation, landfill development, and city planning(box 2). These scenarios deal with potentialchanges in user behavior, the utility of capitalinvestments to a mobile population, and theeconomic and social pressures likely to developon a national scale over time. Special emphasisis also placed on likely voter pressures, largerpolitical and demographic trends, and ways toensure continued political support for projectsonce they are underway. Scenarios can helpcreate the best long-term strategies for contin-ued growth and flexibility for states andregions.

Scenario Analysis for Long-RangeNational Strategy

Significant scenario development also has beendone in the public sector at the national policylevel. Scenarios have been used to shapeagendas for change in Colombia, Guatemala,and South Africa (box 3). In these countries,which were mired in turmoil or facing funda-mental socioeconomic crises, leaders fromacademia, business, trade unions, the military,rebel groups, political parties, and peasantunions came together to discuss steps forward.

When used in the public sector, strategicscenarios have the goal of helping decision-

In the mid-1990s, the State of New Jersey used aparticipatory scenario process to develop a vision forthe future that would help it deal with intransigenttransportation issues. The core decision waswhether to focus on sustaining existing roads,expanding the highway system, or building more lightrail. To develop the scenarios, the scenario teamlooked at the likelihood of regional and nationaleconomic growth, citizen priorities, environmentalpressures and political will. Research, whichincluded a values survey of a cross-section ofcitizens and businesses, studies of existing traveland commerce patterns, and a study of conventionalforecasts, was carried out as part of the process. Inaddition, consultations were held in communitiesacross the state.

The scenario team included state politicians,department of transportation decisionmakers,

experts, academicians, and activists. The scenariosfocused on economic growth and citizen attitudes(see box 7 for full scenario matrix). Special attentionwas paid to regional growth, because New Jersey is amajor transportation corridor between the two largemetropolitan areas of New York and Philadelphia.

The scenarios were widely publicized andsparked significant public debate. All of the sce-narios highlighted the importance of innovation andcooperation with the public sector and the need forenvironmentally sound development. Throughinnovative public-private partnerships, the depart-ment of transportation leveraged federal funds for itslong-term plan of mixed road-rail development.Business practices were also changed to allow formore citizen interaction.

Source: Adapted from Bonnet, 308-24

Box 2. Scenario Analysis for the Public Sector: New Jersey Department of Transportation

makers and key stakeholders agree on substan-tive messages about possibilities for futurechange. The process helps create informalnetworks and understanding among theparticipants that can lead them to becomechampions of the policy recommendations thatcome out of scenario discussions. The secondfunction of public sector scenarios is to engagepublic opinion. The completed scenarios aresimplified and broadcast in the mass media,explained at local meetings, and highlighted inleaflets and policy bulletins. Shaping publicdebate about issues of national concern is madeeasier by the flexibility of the scenario method-ology. Scenarios are stories that bring togetherthe key elements of policy in a narrative thatcan be understood by the public. Because eachscenario can illustrate plausible changes alongmany dimensions, discussion begins to focuson the interactions among forces, rather than onspecific issues.

Adapting Scenario Analysisto World Bank Interventions

In the World Bank, scenarios can provideassistance in formulating goals, designingprojects, building ownership, and monitoringprogress. Scenarios are most valuable whenthey are brought in at the point in the projectcycle at which the overall direction of the

66

Jonathan N. Maack

intervention is being designed or significantlyrethought. If it is meant to be the key strategicframework for an operation, the scenarioprocess should be integrated with the budget-ing, negotiation, and implementation of aproject rather than treated as a stand-aloneexercise.

Such an application of scenario analysiswould add significant value in five strategicareas:

1. Formulation of Country Assistance Strategies.The CAS is the Bank’s business plan in anation, the closest analog to the corporatestrategies around which scenario analysistechniques were developed. Scenarios canincrease the selectivity and realism of a CAS byfocusing discussion on likely future opportuni-ties and constraints. The technique has beenpresented in clinics for Bank task managers onformulating CAS documents. The combinationof a focus on areas of key uncertainty and astrong participatory element in the scenario

analysis make it well suited to this approach.Scenarios are also valuable for monitoring theprogress of both the client and the Bank towardthe goals outlined in the CAS.

2. Formulation of participatory nationalstrategies. The emphasis on participationmakes scenarios useful for planning pro-cesses that need dialogue and agreementamong multiple stakeholders, such as thePoverty Reduction Strategy Paper (PRSP)and the Comprehensive Development Frame-work (CDF). A country-driven scenarioprocess is an excellent way to develop owner-ship and incorporate the political situationand relevant external forces in PRSP and CDFdocuments. Because the scenario process canbe led by the country, as it was in Colombiaand South Africa, the Bank can concentrateon its comparative advantage: leveraging theinput of macroeconomic, political, andsectoral experts to add depth and believabil-ity to the scenarios.

In 1990, four distinct—and evocatively titled—scenarios for South Africa’s post-apartheid futurewere developed during three 3-day workshops. Theprocess was initiated by a local university and tookplace during the heated national debates betweenthe National Party and the African National Con-gress (ANC). The participants looked at the possibleactions of various stakeholders in the politicalprocess and came up with four scenarios.

1. Ostrich. A closed transition process in whichconservative parties attempt to forge a nonrepre-sentative compromise government rather thanreach a negotiated settlement. The outcomes ofthis policy, named after the bird’s habit of stickingits head in the sand in the face of danger, arepolitical and economic isolation and deepeningethnic tensions.

2. Lame duck. A half-hearted and prolongedtransition of power that creates a coalition govern-ment stripped of any real authority. The outcomes ofthis scenario are a deteriorating political climate,lack of sufficient investment in economic redevelop-ment, and the risk of continued isolation.

3. Icarus. A populist national government takespower and adopts an unsustainable policy of highgovernment spending and taxation. The outcomeof this policy is short-term prosperity followed by a

deep economic slump that leaves the country inworse condition than it was during the transition.

4. Flight of the flamingos. A smooth transitionfollowed by a move toward sustainable policies toincrease growth and inclusion. This story, namedafter the slow take-offs and close flight patterns offlamingos, shows steady economic growth, biparti-san priority setting, and a greater increase in long-term benefits to the poor.

The scenarios were developed by economists,businesspeople, politicians from all major parties,academicians, and NGOs. They received heavypress coverage, and F. W. de Klerk, leader of theNational Party, was quoted as saying that he was“not an Ostrich.” The Icarus scenario, which critiquedunsustainable government spending, was endorsedby key left-wing economists and comforted the localand international business community. The transitiongovernment that eventually came into power wasnotable for its restrained economic policies and itssuccess at using tested, sustainable methods toreach the poor. The scenario process was anexcellent way to broaden the involvement of stake-holders in policy dialogue without creating chaos inthe decisionmaking process.

Source: Adapted from Kahane, 325-32.

Box 3. Scenario Analysis for National Strategy: South Africa

67


3. Formulation of a medium-term framework forprogram loans. Scenario analysis is most effec-tive in looking at medium- to long-term timehorizons. It is thus well suited to helping theteam prepare a medium-term framework thatlooks at the desired outcomes of lending andthe uncertainties that it will face. Because thesignposts of progress toward different sce-narios are formulated by a group, the processmakes the framework more transparent andincreases the motivation of both Bank andclient to act on strategic recommendations.Scenarios also aid in the development of exitstrategies based on mutually understoodchanges in the agreed indicators.

4. Developing benchmarks and goals for anAdaptable Program Loan. Scenario analysis canprovide an effective and inclusive method tocreate a long-term set of benchmarks forextended interventions, such as the AdaptableProgram Loan (APL). By including all stake-holder representatives in the risk analysis andindicator development process, scenariosestablish a clear and consistent set of bench-marks that are widely understood. Due to theneed for regular review and approval in theface of emerging needs, the APL project cyclefavors the regular monitoring and revision thatmake scenarios most useful.

5. Analysis of risks and impacts of adjustmentloans. Adjustment operations frequently in-volve institutional change. Scenarios are anexcellent way to look at the potential outcomesof changing the formal and informal rules in asector or an industry. Changes in the waypeople do business can have unexpectedresults. Disciplined scenario analysis enablesdecisionmakers in the Bank and the clientgovernment to clarify the drivers of change in agiven sector and explore potential interactionsamong those drivers and the impacts ofchanges. Because they build consensus amonglabor, public, and private sector leaders aboutwhat needs to be changed and why, scenarioscan help task managers shape the publicdialogue about adjustment operations byclarifying the objectives and reasoning behindan intervention.

Adapting and Shorteningthe Scenario Process

Scenarios can be adapted to enhance strategicplanning at a number of stages in the projectcycle. Scenarios are, above all, a way of thinkingabout the world and can be used in shortenedform to change the terms of dialogue and intro-duce a new way of interacting. A brief, participa-tory scenario workshop can be arranged early inthe design phase to build scenarios and gatherinput quickly. Alternately, a more extensiveprocess with supporting research can be carriedout during the development of a risk manage-ment framework and used to develop a systemfor monitoring and evaluation.

An example of a streamlined scenarioimplementation may clarify how the techniquecan be adapted. If the task manager of anadjustment loan wanted to use scenarios as oneinput during project design, but did not wantto execute the full process, he or she could setup scenario workshops with relevant stake-holders during preparation and appraisal tobuild simple models of relevant forces. Thesebasic scenarios could be sketched out by thegroup in two one-day workshops during a two-week preparation mission as a form of brain-storming and negotiation. They could then beintegrated in the early stages of project plan-ning. If there were enough time and money, theperiod between preparation and appraisalcould be used to flesh out the scenarios withstatistical projections and other relevant infor-mation. The final workshop, in which thecompleted scenarios would be further ques-tioned by the group and applied to createcontingency strategies, could be held duringthe appraisal mission.

Running a Scenario Exercise

The following description of the stages of ascenario exercise will give task managers abackground in its practical requirements. Thisdescription should enable task managers tolead a simplified scenario analysis process,make a significant contribution to a scenarioteam, or work knowledgeably with privatesector consultants.

68

Jonathan N. Maack

Stages of the Scenario Process

There are no fixed rules for how to carry out ascenario analysis, but several stages occur inalmost every case. The actual creation ofscenarios takes place with a group in a work-shop. The organizers need to carry out anumber of activities before the first and lastworkshops. The four broad stages to theprocess follow. The specifics of each stage willbe discussed later in the chapter. An illustrationof a sample process from start to finish can befound in appendix 2.

1. Preparation. A number of tasks need to becarried out before the scenario workshops. Theyinclude defining the scope of scenario use,choosing and interviewing the scenario team,and preparing background material for theworkshops. These are crucial steps in the process,because they help the task manager build confi-dence in the scenario exercise and gain an earlyunderstanding of the issues that will be mostimportant to participants. The specifics of thisprocess are detailed in appendix 1.

2. Building the scenarios. This process canoccur in one workshop, or two held back toback. Together the scenario team determinesthe focus of the intervention, identifies andprioritizes the key external factors that willaffect its success, sets them in a matrix todifferentiate their effects, and writes simplescenario plots.

3. Investigating and writing the scenarios. Thisexercise is carried out between workshops. Thetask manager and core team gather relevantqualitative and quantitative information todetermine whether the assumptions made inthe scenarios are accurate. Decisionmakersfrom the scenario team should write morefinished drafts of the scenarios.

4. Using the scenarios to plan strategy. In thisworkshop, the scenario team goes over thefinished scenarios and looks at the implicationsfor the strategic decision being made. Specificexternal “signposts” of changes in the keyforces are also identified at this stage.

Building the Scenarios

The first step of the scenario workshop is toclarify the decision focus of the exercise. Thisshould have been outlined to individuals whenthey were invited to participate, but it isimportant for the group to work together toclarify the issue. As this discussion can becomecontentious, it is best to state clearly within thefirst 10 minutes of the meeting the behaviorexpected during the meeting. Propose, or askthe group to agree on, whether a facilitator isneeded, and, if so, who it will be; a system torecognize speakers; a time limit on the lengthof comments; and other basic procedures of themeeting.

Discussion of the decision focus forces allparticipants to concentrate on the goals soughtby the development intervention rather than onthe specific process that will be used to reachthose goals. This separation of means from objec-tives is key to giving the group the conceptualspace needed to come up with new approaches.Several key issues should be addressed whendeciding the decision focus:

1. Desired outcome of the intervention. Thisdiscussion should begin with input from theorganizers of the event but then will be open tofurther discussion. The group may come upwith a different formulation of the desiredoutcome than was originally brought to thetable. This reformulation will guide the group’sunderstanding of the forces that will affect it.

2. Desired outcome of the scenario process. Oncethe outcome of the intervention is clarified, therole that scenarios will play should be ex-plained. Will they be used as a way to testexisting strategies for riskiness, as a tool toformulate new strategies, or as a means tobuild a shared vision for governmental ororganizational reform? The organizers shouldclarify how they envision the scenarios beingapplied to strategy, and explain any internalbusiness tools that will be used to put thatstrategy into practice. For example, if the goalis the creation of a medium-term framework toguide the sequencing of a program of adjust-ment loans, the group needs to understand

69


how the program of loans works and wherescenarios fit in.

3. Time frame. The time frame will signifi-cantly affect the range of issues a scenario canaddress. The technological and economic shiftsthat should be considered in a 20-year timehorizon may have no relevance for a 5-yearplan. Scenarios should be conceived of as waysto shape strategy within the medium term. Thetime frame should be long enough for signifi-cant changes to occur in the driving forces,ideally 3 to 5 years.

Key Factors

After the decision focus has been discussed andanalyzed, the team should brainstorm keyfactors for the success of the strategy or opera-tion. This session should be led by a personwith experience in facilitation. Key factors forproject success can include:

• Specific events, for example, the passage oflegislation that gives legal status to NGOmicrocredit providers

• General trends, for example, a decline in thecost of fertilizer

€ Improvements in service delivery measuredby positive beneficiary feedback as well asnumber of persons served.

To get as many dimensions of success aspossible, the facilitator should encourage allmembers of the group to participate duringthis phase. After a brief brainstorming session,the clusters of logical association and keyfactors developed by the analysts after thepreliminary interviews (see appendix 1) can beadded to the list. These can become buildingblocks for discussion and make it easier fordifferent groups to understand one another.

Two Driving Forces: Externaland Organizational

The free association used to identify key factorscan lead to a jumbled list. The next phase in theprocess consists of the group’s looking at theunderlying causes of the issues that have comeup in individual interviews and group brain-

storming. The goal of this stage is to build agood conceptual model of the forces that aremost relevant to the decision focus.

There are two large categories of drivingforces. External forces are the social, economic,environmental, and political forces in societythat are relevant to the topic of the scenariodiscussion but are outside the participants’control. Organizational forces are the product ofthe actions of the Bank and other stakeholdersand can be affected by the project. These driv-ing forces will become the key foci of furtherscenario development, and they can be identi-fied intuitively by using the visual clusteringmethod described in appendix 1 or separatedinto a series of logical groups and analyzed.

External Forces: The SEEPT (Social,Economic,Environmental, Political andTechnological) Framework

The simplest way to systematically addressexternal forces is to organize the issues identi-fied in the brainstorming sessions and prelimi-nary interviews in five broad categories:

1. Social2. Economic3. Environmental4. Political5. Technological.

The SEEPT framework (table 1) is an efficientway to obtain a holistic view of the many forcesthat will affect a project’s success. Rather thanfocusing on a single dimension, such as eco-nomic performance, the group can look at howthe forces relate to and affect one another. Thespecific events identified in the initial brain-storming of key factors can be thought of assymptoms resulting from these deeper forces.For example, the strain on urban infrastructurethat results from migration to cities is inti-mately related to demographic, social, andeconomic pressures. It may also be tied topolitical pressures or regional instability. Asimple device like the SEEPT framework canhelp decisionmakers look at all the dimensionsthat affect an urban development project atonce. As the group thinks about how theseforces could plausibly change and interact in

70

Jonathan N. Maack

Table 1. Illustration of the SEEPT (Social, Economic, Environmental, Political and Technological) Framework for Identifying External Forces This is an illustrative SEEPT framework for a national-level scenario exercise. This chart can serve as a menu for scenario teams or as a stand-alone example of the sort of questions that can be posed in each of the SEEPT categories. The Social Indicators of Development, Country Political and Institutional Assessment information, United Nations Development Programme (UNDP) Human Development Indicators, and the World Development Indicators are useful sources of data for these domains.

Domain Category Examples Social Social factors Education levels, social priorities, cultural and class tensions, land and

water rights, differentiated membership in groups and associations, gender issues

Demographic patterns Age, family, household, and ethnic structures; regional and national migration patterns; wealth distribution, including regional and national poverty rates

Economic Macroeconomic conditions

GNP, balance of trade, rate of inflation, exchange rate; current and future relations with international financial markets, current debt levels; governmental expenditures, deficits Changes in the economic structure of nation (dependence on single export, percentage of exports in finished goods), formation of new regional trading blocs

Microeconomic conditions

Change in size, type and ownership of firms; formal and informal labor force structure by region; changes in economies of scale/structure of key industries

Market forces Spending patterns of consumers (urban/rural, national/regional), international demand for key exports Distribution and efficiency of rural and urban markets, impact of the informal sector, sources of competition (national, regional, international)

Impacts of global economy and development

Volume of assistance from multilateral and bilateral agencies, conditions for assistance (policies, requirements)/harmonization Risk tolerance and conditions for entry and exit by international firms, stake in local economy by international firms

Environmental Physical environment Air/water/land pollution trends and locations, environmental quality issues (global warming)

Natural resources Energy prices and availability (likelihood/impact of an oil shock), raw materials (rate of depletion, ease of access), land use (farming methods, erosion levels), sustainability (strategic use of resources) regional distribution of natural resources

Political Geopolitical Trends in international relations; relationship with other nations in region (regional trading blocs, military alliances); levels of tension, conflict (regional, international); trade and protectionism

National Change in governmental development strategy and policy (privatization); changes in legislation (including regulation, creation of enabling environment); changes in structure and responsibility of ministries; changes in rules governing formation and functioning of parties; stability of government, likelihood of change/overthrow

Technological Infrastructure Level of technology in key industries, emerging technologies, capacity to manufacture technology for export

Future directions Basic research and technical education trends in nation; “digital divide” –computer and telecom infrastructure/trends; potential for the rapid diffusion of new technologies from abroad

Source: Adapted from Wilson 1987.

71


the future, the group could create differentiatedstories about how broad future trends anddevelopments could affect the operation.

Organizational Forces

Organizational forces are the specific actionsand general positions taken by actors (indi-viduals, agencies, and organizations) that affectproject success. These actors may includegovernment agencies, the private sector,political parties, NGOs, and internationalagencies. General trends in governmentalactions might include a move towardprivatization of state industries or trade liberal-ization. As in the case of external forces, thebehavior of actors is interconnected. If thegovernment and trade unions work together tocreate an enabling environment for trade,international businesses might begin to makeserious investments or solidify their presence.The actions of agencies and organizationsshould make sense within the economic andpolitical environment of a scenario. If organiza-tional forces are used as key scenario drivers,they should be consistent with past behavior inthat organization. Radical changes of positionon critical issues should not be incorporatedunless the rest of the scenario environmentmakes such changes likely.

Prioritizing and Analyzing External Forces

Identifying key factors and assembling a list ofrelevant forces are the first steps. This list canyield as many as 50 driving forces and keyfactors. The next steps are analysis andprioritization, which reduce this larger set offorces to those that are most relevant to thedecision focus, are predictable vs. unpredict-able, and will respond to actions by participat-ing agencies and groups. Complex analysis isnot required. This is a logical exercise based onrigorously thinking through the forces andtrends already noted and identifying the onesthat are most important for the decision. At aminimum, this discussion should include:

Identification of cause-and-effect relationships.How do these forces interact? What impact dothey have on other forces?

Apparent direction of these forces today. Whatare the current trends in these forces andwhy? One of the most critical tasks for sce-nario analysis is a good understanding of thecauses of present trends and conditions.Present trends are a good base from which toextrapolate future interactions among drivingforces.

Relevance to the decision focus. What is themagnitude of the impact of these forces on thefuture course of key factors for project success?Are reforms dependent on a political coalitionthat draws its support from a constituencywhose influence is shrinking?

Future prospects. How much, in what ways,and how fast might these trends change in thefuture? At this stage of analysis it is integralthat the scenario team differentiate amongtrends that are:

• Predictable. Certain things, such as currentdemographic trends, are already known andwill need to be taken into account in allplanning.

• Unpredictable/uncertain. Market prices,currency rate fluctuations, demand forexport goods, and changes in political valuesare all unpredictable elements.

The impact-uncertainty matrix is an easyway to combine key factors and drivingforces in a clear, legible format. The matrixranks the forces on a simple “high-medium-low” rating system. The best way to deter-mine prioritization is through a vote. A votegives all participants equal voice and leadsto a picture of the key forces that incorpo-rates multiple viewpoints. The facilitatorshould give each person in the group ap-proximately 20 chits to “spend” as “votes”and ask each individual to grade each forceaccording to:

• Its impact on the key decision factors (aparticipant should “spend” more chits on theforces he or she sees as most important) andthe degree of uncertainty (each participantshould “spend” more chits on the forces thatare uncertain)

72

Jonathan N. Maack

• The number of chits that equal a “high,”medium,” or “low” rating should bedecided on by the group based on thenumber of forces that the group has devisedand the number of people in the group.

The impact/uncertainty matrix (table 2) is anillustration of how the ranking system can beused to narrow the list of forces to the mostrelevant for differentiating scenarios. Scenarioplots should be built around high-impact/low-uncertainty issues (highly relevant issues withpredictable future outcomes for which currentplanning must prepare) and high-impact/high-uncertainty issues (issues that could shapedifferent futures which planning should takeinto account).

Creating a Scenario Matrix to Outline FourDistinct Scenarios

The results of the impact/uncertainty matrixare seldom conclusive. A number of forces willcluster in the upper right hand corner (highlyunpredictable/highly relevant). The groupneeds to pick the two most relevant of these forcesand set up a scenario matrix using them as theaxes. The group will then work out the fourpermutations of these forces and build thescenarios around them.

The scenario matrix (box 4) ensures that eachscenario will be different in a logical, nonran-dom way, because the top scoring factors willbe key drivers in all scenarios. The scenarioteam discusses each permutation of the two

Table 2. Example of Impact/Uncertainty Matrix

Degree of uncertainty

Low Medium High

Critical planning

issues Highly relevant and fairly predictable (can often be based on existing projections). Should be taken into account in all scenarios.

Important scenario

drivers Extremely important and fairly certain. Should be used to differentiate scenarios. Should be based on projections but potential discon-tinuities also should be investigated.

Critical scenario

drivers Factors and forces essential for success and highly unpredictable. Should be used to differentiate scenario plots and trigger exit strategies.

High

Important planning

issues Relevant and very predictable. Should be figured into most scenarios.

Important planning

issues Relevant and somewhat predictable. Should be present in most scenarios.

Important scenario

drivers Relevant issues that are highly uncertain. Plausible, significant shifts in these forces should be used to differentiate scenario plots.

Med

Monitorable issues

Related to the decision focus but not critical. Should be compared to projections as scenario is implemented.

Monitorable issues

Related but not crucial to the decision focus. Should be monitored for unexpected changes.

Issues to monitor and

reassess impact Highly unpredictable forces that do not have an immediate impact on the decision focus. Should be closely monitored.

Low

Level o

f imp

act

Note: Shaded areas indicate key focus. Source: Adapted from Wilson 1989.

Table 2. Example of Impact/Uncertainty Matrix

Note: Shaded areas indicate key focus.Source: Adapted from Wilson 1989.

73


most relevant forces. The decisions made abouthow these forces are likely to behave sets thetone for each of the four different scenarios.This process enables the quick formulation ofdistinct, logical scenarios. Because the decisionon the two axes of uncertainty is group-driven,it encourages group members to challengeone another and seek justification that acertain force is the best way to differentiate ascenario. Narrowing the discussion enablesthe group to explore ways in which these twoforces could develop and how these develop-ments can help focus the other forces arounda distinct plot.

What Makes a Good Scenario Plot?

The scenario matrix approach is designed tocut through many complex issues and isolatetwo sources of great uncertainty and greatimportance for the success and sustainability ofthe work being planned. This process of reduc-tion may leave members of the scenario teamuncomfortable, because not all issues can beeffectively addressed in terms of these two keyaxes of uncertainty. Although these forces willremain the skeleton of each future narrative,the next phase involves the addition of detail tothe stories and the inclusion of a larger numberof driving forces. A good plot combines drivingforces in a dynamic, coherent story.

Key Elements

The key elements of a good scenario plot are:

• “Critical Planning Issues” from the Impact/Uncertainty Matrix. These are the compara-tive sureties in the future that are particu-larly relevant to the decision focus. Thecritical planning issues may concern specificaspects of the issue that the task team isstudying and should have a role in all of thestories about the future. Issues such asdemographic trends should be implicit in allplots, but the role they play may be quitedifferent depending on how political, social,and economic factors affect issues such aseducation, employment, emigration, andconsumption.

• “Critical Scenario Drivers” from the Impact/Uncertainty Matrix. Each of these factorsshould play a role in at least one scenario.That role should work naturally with theother forces in play in each scenario, addingthe twist of giving decisionmakers a chanceto explore how a particularly positive ornegative development might affect theoperation/strategy as a whole.

• Systems of interaction among forces. Thenumerous relevant forces and issues devel-oped by the group should be exploredsystematically in each of the different sce-narios. Changes in the two key axes ofuncertainty can be thought of as stonestossed in a pond: their aftereffects will ripplethrough the host of other forces that alsobear strongly on the decision focus.

An easy way to start the process is to thinkof different static “end states” that could resultfrom the interaction of two forces, and workbackwards to explore how forces would needto interact to reach that point. Some of the keyfactors cited by the group earlier in the exercisewill be precisely this sort of “end state.” It isimportant to remember, however, that sce-narios are not descriptions of a static futureworld described in great detail. They arenarratives about how events might unfoldbetween now and a future date given the waysin which the forces that the scenario team hasidentified interact.

Scenarios should be:

1. Plausible. The events in the story should bepossible, and the narrative should be credible(the descriptions of what happened, and whyand how it happened should be believable).

2. Distinctive. Each scenario should focus on adifferent combination of the key forces. Sce-narios should be clearly differentiated instructure and in message, not variations on asingle theme. Multiple scenarios should be usedto explore how different permutations of thesame key forces can yield very different worlds.

3. Consistent. Each scenario should have astrong internal logic. The goal of scenarios is to

74

Jonathan N. Maack

The decision focus was whether transportationinvestment should focus on sustaining existinginfrastructure, expansion of highways, or thedevelopment of light rail. The group agreed on thethree most important driving forces for the sce-narios:

1. Rate of regional and national economic growth.The team looked at high, medium, and low levels ofGNP growth and the rise or fall of the two majorurban areas that affected the state. They studied thenature of growth (technological or industrial) and theeffect of the economy on private investment intransportation and public sector revenue.

2. Values and attitudes of citizens. Citizens weremodeled as inwardly oriented and individualistic, oroutwardly oriented and communal. The inward-looking citizen had a continuing love of cars andsingle-family homes, a strong distrust of government,and was rent-seeking. The outwardly oriented citizen

was more urban and environmentally conscious,wanted to revitalize cities, and trusted governmentmore.

3. Government leadership. The government wasmodeled as oriented to the short term and crisismanagement, or oriented to the long term and ableto stay with a stable plan.

Each of the scenarios looked at the relationshipsamong the three driving forces and their effects onfactors such as the environment, investment,education (what is the workforce doing?), and trafficflows. Scenario logics were consistent. A low-growthscenario had a greater split between rich and poor,an emphasis on safety, and a government thatfocused on short-term solutions due to constrainedresources. In another scenario, technology drovestrong national and regional growth; immigrants weremiddle-class; and convenient transportation wasemphasized in more densely developed urban areas.

Box 4. Scenario Drivers in the New Jersey Department of Transportation

Scenario Matrix: New Jersey Department of Transportation

High Growth Gateway to the world: Pushing the envelope: A high-tech future A sustainable future · Demand for efficiency, strong leaders · Movement to urban areas · Modest GNP growth · Strong public transportation, good leaders · Education matches skills needed · Economic growth based on · Private/public sector invests in highway transportation

sustainable technologies Inward-looking Citizens Outward-looking Citizens

Bad News: Worst Muddling through: possible future Status quo continues

· Class and race hostilities grow · Investment decisions are made politically · Public distrust of government · Public transportation is seen as important

· Decline of New York City hurts NJ but only partially funded · Growth slow, high unemployment · GNP grows slowly, low public spending · Large gap between rich and poor · Gap between rich and poor grows

Low Growth

Source: Adapted from Bonnet, 308-24.

75


explore the way that forces interact, and eachaction should have a reaction. Neither actorsnor external factors should completely over-turn the evidence of current trends and posi-tions unless logical explanations for thosechanges are a central part of the plot. Forexample, it is highly unlikely that there will below inflation and high growth, or that a regula-tory agency that was formerly very strict willsignificantly loosen its requirements withoutsome extenuating circumstances.

4. Relevant. Each scenario, and the entire setof scenarios taken as a whole, should contrib-ute specific insights into the future that relateto the decision focus chosen by the group. It isimportant for the author to return frequently tothe objectives of the exercise as she or he buildsthe scenario and assesses which of the externaland institutional factors created in brainstorm-ing should be explored.

5. Creative. The emphasis on logic and rigorin the process should not stamp out innovativethinking. Part of writing a scenario is enticingthe reader into the scenario world. The authorcan do this by telling his or her story throughparables, citing examples, or using famousfigures as stand-ins for complex ideas.

6. Challenging. Scenarios should challengeconventional wisdom about the future. Whenthinking about the major sources of uncer-tainty, the author should try to explore alterna-tives that will significantly alter the basis forthe assumptions that go into the “officialfuture.” It is often worthwhile to write a “wildcard” scenario that uses some of the key forcesnot listed in the scenario matrix. This gives theset of scenarios as a whole more breadth andstrength.

Scenario Plot Types

Scenario plots should differ from one anotherwithout being wildly positive or negative. It isimportant that they challenge assumptionswhile remaining balanced enough that they arenot dismissed out of hand. Here are some plot“types” that can help create stories about thefuture:

• Winners and losers. These scenarios explorethe future in terms of ascendant versusdeclining organizations, nations, or socialforces. Examples include scenarios that lookat trading blocs, focusing on dominantgroups and exploring how different nationswould fare in a climate of struggle andchange.

• Good news/bad news. These scenarios uselarge external factors, such as currencydevaluation, a massive increase in corrup-tion, or breakdown of the rule of law, toencourage rethinking existing strategies andpriorities. These stories ask decisionmakers“what if?” questions about the logic of theirlong-term plans in the face of key forces thatcan move in unexpected directions.

• States of change. This model is one of additivechange, a world in which a series of alter-ations feed off one another to move societyinto a different mode. These alterations caninvolve a series of innovations leading toimprovement, or a series of mistakes leadingto political gridlock and economic stagna-tion. Some models focus on structuralchange in a nation (the ending of apartheidin South Africa), an industry (the introduc-tion of software to produce online reports,removing the need for specialized printers),or a sector (environmental movements spurstrong pressure to build public transporta-tion and reduce pollution) and how thechange will affect different actors and theviability of investments.

• Cycles. Another way of looking at change isthrough cycles, both economic and political.In this model, the rise and fall of socialgroups or stakeholders can be charted inrelation to the economic cycles of relevantindustries or international trends.

• Wild cards. Wild-card stories explore howcatalytic developments could completelyreshape a society. For example, the develop-ment of the internet and its effect on busi-ness in the United States is a wild-cardscenario that is still being played out. Otherexamples might include the unexpecteddiscovery of oil reserves or other resources.An example of this is Tuvalu, an impover-ished South Pacific island nation, whichdiscovered it had a unique asset. Tuvalu’s

76

Jonathan N. Maack

internationally designated internet countrydomain is “.tv.” Tuvalu has sold the rights toits domain name to a Canadian companythat will market it to television and radiowebsites. This sale will earn the 10,000Tuvalu islanders as much as $50 million overthe next 5 years.

Investigating and Writing the Scenarios

A thorough investigation of the trends underly-ing the scenario plots is essential to writingcredible scenarios. If scenarios are not based onsignificant quantitative data and a carefulreview of a country’s political and culturalsituation, the scenario process amounts to abrainstorming session. Writing the final sce-narios should include a process of triangulationwhereby the ideas and principles discussed inthe workshops are investigated through origi-nal research and consultations with experts.Qualitative and quantitative research onrelevant issues should be carried out before thescenarios are written. The former is based oninterviews and follow-up questions withparticipants and key stakeholders identified inthe scenario meetings. The latter includeseconomic projects, financial sustainabilityresearch, demographic studies, and statisticalanalysis.

Key research tasks follow:

• Interview experts. Speak with academics,sectoral and regional experts in the Bank,and other specialists on the best way toaddress technology, the political situation,social issues, and trade and economics.Interviewing is an excellent first step be-cause interviews can be combined with thescenario team’s requests for help in identify-ing the best way to quickly gain an under-standing of a subject or the nation.

• Assemble and disaggregate existing information.These steps involve reading relevant storiesin the local press, conducting quick literaturereviews on key subjects, collecting bestpractice cases, and assembling quantitativeinformation on the predictable and relevantforces in the scenario.

• Synthesizing quantitative work. Projections oftrends for all of the key forces should be setup on a single chart. How do they interact inexisting projects? Do they move together?How does this “official future” compare withthat shown in the scenarios? Comparenational, regional, and international trendsalong the driving forces. If possible, eco-nomic models for the predicted forcesshould be created and looked at for howthey compare to past data.

• Carrying out original qualitative work. Thismay involve conducting stakeholder inter-views to analyze the strength and impor-tance of political or cultural tensions,working with local universities to createprofiles of likely user groups, or other fieldwork. Other tasks might include disaggre-gating household surveys, and analyzing thegeographic distribution of the poor and bestpractice projects.

Writing Full-length Drafts of Each Scenario

Once the scenario plots have been sketched outby the whole scenario team in a workshop,they should be thoroughly researched andwritten up as full-length drafts. These draftsshould be written by key decisionmakers,ideally the task manager and a highly placedgovernment official in the client country. It isessential that the authors be present for theentire scenario discussion and have a clearstake in developing and implementing effectivescenarios. Decisionmakers will be asked to shapethe strategy for a project or a nation using thesescenarios, and the best way to ensure that they own,understand, and believe in them is to give themauthorial control.

A scenario should have a creative title thatsums up its central message. It is best to beginthe scenario with a brief summation thatexplains the title and sums up the elements ofthe story agreed in the workshop. The finisheddrafts of the scenarios should be smoothnarratives that are easy to read and under-stand. Each scenario will be built around aseries of specific events, but the underlyingforces and their interactions also should be

77


made clear. This phase of writing involvesinterweaving the social, economic, political,and technological scenario drivers into asmoother whole.

The scenario in box 5 was created by ascenario team funded by the Colombian pri-vate sector and led by private consultants. Itreflects the input of economists, politicians,academicians, military officials, guerilla lead-ers, church officials, and peasant and studentleaders. It is a good example of a smoothnarrative form for a complex story. It movesfrom economic events to political and socialissues and explores the way that forces influ-ence one another over time. The entire scenariois written in the first person plural (we) toimmediately connect the reader to the eventsand decisions being described. It works chro-nologically, highlighting the impact of a failureof leadership along economic, political, andsocial lines as the country moves from bad toworse.

Box 5. Destino Colombia Scenario: “When the Sun Rises, We’ll See”

Weariness, laziness or inability to face problems are all justified by the phrase “When the sunrises, we’ll see.” Darkness turns into a pretext for dreams and apathy, but the clear light of dawn,rather than inspiring important decisions, simply gives rise to a new period of uncertainty. In theface of the country’s crisis, this irrational confidence in unexpected, miraculous outcomes, thisrecourse to halfway solutions, this generalized tendency to put off basic actions until later, havecombined to the point of becoming a collective alternative. Before its enormous and rapidlygrowing problems, Colombia seems to be saying “When the sun rises, we’ll see.’”

— Quote from the Destino Colombia scenarios

As crises repeatedly rock the country, exhaustion seeps into the government and the populace. Continualcrises lead to a focus on solving problems day to day. The government wastes valuable time and energynegotiating without a strategy and is captured by pressure groups including the wealthy, various regionalwarlords, and corrupt officials. This “law of the strongest” leads to the extraction of the greatest costs from themost successful industries, the growth of the drug trade, the rise of regional governments with their owntaxation systems, and the end of a national oil industry and economic independence.

As the situation deteriorates, the poor suffer the most. International organizations step in but are able toprovide only stopgap solutions. Ultimately, increasing human rights violations and economic disorganizationlead to the country’s becoming economically and politically isolated. This isolation leads to military threats tothe borders by neighboring countries and increasing levels of emigration. The scenario closes with thecomment that the disintegration was slow enough that people simply “became numb to the proliferation oftheir misfortunes.”

Source: Adapted from Kahane, 11-14.

Using Scenarios to MakeStrategic Decisions

The final task of the scenario team is to explorethe relationships among the plausible futuresthey created and the strategic decision at hand.The strategic options presented by each sce-nario should be discussed in a workshop withthe whole team present. This process enablesdecisionmakers to reap the full benefits ofgroup involvement in the process. Whenlooking at each scenario, the scenario teamshould take the following actions:

1. Step into the narrative. This involves re-reading the scenarios and questioning them asecond time. The team is thereby able to re-engage intellectually and emotionally with thematerial. The team members should envisionthemselves in this future world and thinkabout how the conditions that define it wouldchange their professional and personal lives.

78

Jonathan N. Maack

2. What are the implications of this world? Eachscenario provides a different world for theintervention. This step involves a discussion ofthe implications of each scenario for the centralstrategic decisions that need to be made. Astandard set of questions designed to apply tothese decisions should be created for this phaseof discussion and used for each scenario. Ateam might want to look at which ministriesare most affected by each scenario or whichgeographic areas would experience the mostchange.

3. What is the best strategy for this scenario?Each scenario presents risks and opportunities.The team should focus on the broader strategiccontext and answer the following questions:

• What is the best strategy for dealing with thissituation? The scenario team should look firstat how the intervention can most effectivelymove toward its desired outcome in thisfuture world. What general strategic positionshould be taken? Should it be one of inter-vention in certain areas, or restraint to allowthe situation to run its course?

• What are the major opportunities and threats in thisscenario? This may involve looking at howdifferent sectors will be affected by the changesoutlined in a scenario. It may also involvelooking at the effects that different scenarioshave on specific social groups, identifying thewinners and losers in each story.

• If we know this scenario will come to pass, theBank/client should (name one action)? It shouldstop doing (name one action)? These questionsconcretize the discussion by relating it to thespecific set of actions that are being per-formed (or proposed) as part of the interven-tion. In addition, the scenario team shoulddevelop exit strategies for all or part of theintervention based on each scenario. Whatsituations will trigger the abandonment of aproject or set of projects?

By answering these questions, the scenarioteam has in effect developed a series of simplecontingency plans for each potential future.The next phase of the process is to assess howmuch these strategies have in common with the

current strategy of the Bank and the client. Inthe case of a specific loan, this encourages thescenario team members to look at how whatthey have learned about likely impacts on theintervention fits into the national developmentstrategy.

When scenarios are designed to create aframework for a series of loans, it is moreappropriate to compare the conclusionsreached to the consensus about development inthe country. This consensus can be identifiedthrough Country Assistance and/or PovertyReduction Strategies as well as through com-parisons with the national strategies of otherdonor agencies.

The team should look at the contingencyplans suggested by the scenario team andidentify:

1. Which alternatives seem to be suggested by amajority or all of the scenarios? These should bekey parts of any strategic plan. They reflect aninteraction of forces that is very relevant to theintervention and that the scenario team feels islikely to occur.

2. Which alternatives challenge most strongly theassumptions underlying the current strategy? Theteam should look to these alternatives forguidance in rethinking strategic orientation.Even if the scenario from which they are drawndoes not come to pass, they highlight a blindspot in current plans. When making finalstrategic decisions, the scenario team shouldlook for creative ways to include these insightsin the intervention.

3. Which alternatives are logical extensions ofthe current strategy? These alternatives givedecisionmakers an idea of how alternate futuredevelopments could be leveraged to pushforward an agenda or program that is alreadyin place.

Each alternative strategy presented by thescenario team should be carefully consideredduring the workshop. The final step in theexercise is deciding which strategic alternativesshould be adopted. Again, a simple set ofquestions can serve as a guide:

79


1. What events would trigger each strategy?What impact (positive or negative) wouldthose events have on the project? How effectiveis the strategy at addressing these issues?

2. What is the evidence to support the assump-tions underlying the strategic suggestion? Whataspect of the scenario serves as the underpin-ning of the strategy?

3. Is it feasible for the organizations involved toexecute the alternative? What would preventthem from being able to do so?

Develop Monitorable Progress Indicatorstoward Each Scenario

Once each scenario has been tied to a set ofstrategic actions, the team must develop a wayto look at what is happening in the country andthe world to determine which scenario iscoming to pass. The changes described indifferent plots (unified vs. disintegrated poli-ties, outbreaks of ethnic violence) do not occurovernight. The scenario team should identifyboth indicators based on fixed events andindicators based on trends.

• Fixed events. Fixed events are pivotal inci-dents that push groups into conflict or createeconomic opportunity. They can be externalor internal events and may be time-sensitive(elections) or ongoing external processes(currency fluctuations due to the collapse ofmarkets in Asia). The group should identifyat least four “turning points” for each sce-nario. Events such as oil prices reaching anall-time high or a 20 percent decline intourism can serve as signs that the world isdeveloping along the lines of the scenario.One way to identify such turning points is toinvent newspaper headlines from the futurethat sum up crucial turning points or sym-bolic indications of the forces at work in thescenario. As these events occur, it is impor-tant to look at how their outcomes are relatedto the scenarios that have been created.

• Trends. Indicators of trends can includemeasurable changes in attitudes, demo-graphic shifts (urbanization), and macroeco-

nomic changes. These indicators should betracked throughout the life of the project andcompared to the scenarios developed at itsoutset to determine progress toward a givenfuture. Monitorable indicators can include(1) economic figures, such as export andgrowth levels; (2) demographic trends, suchas rate of migration to geographic areas; and(3) employment trends, such as the numberand type of jobs created in the formal sector.

• When choosing trends to monitor, scenarioteam members should look at where theycan obtain data, how much it costs to pro-duce that data, and how feasible it is toregularly track it. Another strategy is to lookat how data that is already collected wouldbe influenced by the driving forces of ascenario. By tracing the likely impact of ascenario on an already monitored set ofdemographic or poverty indicators, the teamcan ensure that the information will berecorded in a standardized and reliable way.Thinking about how each scenario wouldtranslate into migration, poverty rates, orother affected data helps make the world thescenario team describes more real.

Make It a Reality

The final stage of the scenario process is thedissemination of the message and its imple-mentation on the ground. These two goals arefundamentally linked. One of the keys tosuccessful implementation is giving the publicand key stakeholders (especially those whowill be carrying out the strategy) a goodunderstanding of the lessons learned from thescenarios and the changes in policy that havebeen made as a result of them.

The basic steps are to:

Simplify scenarios and prepare them for dissemi-nation. Reduce each scenario to a 1- to 2-pagenarrative that can be read or explained to anaudience in 2 minutes. Use newspaper inserts,radio and television broadcasts, and videosshown during clinics and seminars to dissemi-nate the scenarios. In Colombia, a video wasmade with various team members explainingtheir scenario with the aid of cartoons.

80

Jonathan N. Maack

Develop support for the scenarios. Once thescenarios have been written, it is essential toinvolve larger groups of stakeholders to dis-cuss them. During the dissemination period,the scenario team should build relationshipswith local politicians, influential figures, andimportant civil society organizations. The teamcan organize clinics and seminars for clientcountry officials, local political leaders, andother figures and use handouts and videos toexplain the scenarios and the strategy thatresulted from them. It is important to monitorlocal and media responses to the scenarios andadjust for them in presentations and seminars.

Build accountability systems. The scenariosshould lead to the definition of clear goals andpathways to alternate scenarios. It is importantto involve the individuals who will generatethe data in designing the implementation. Theteam should develop incentives and mandatesfor data collection and reporting. Responsibili-ties should be clear, and the relationshiplinking the data collected, the scenarios, andstrategy should be explained. For example,insights from scenarios should be tied to thechoice of baseline survey during the designphase and to the strategy for monitoring andevaluation during supervision.

Implement. Clarify the expected results of theintervention and define a clear strategy. Re-solve residual conflicts between previous waysof doing business and the current system.Target priority areas for strategic resourceallocation. Provide guidance and training(including a background on the scenarios) to allof the line managers who will implement theprogram.

Monitor progress and test results. Tie decisionsabout proceeding with the intervention to theindicators of progress toward certain scenarios.Scenario indicators should be a part of the Mid-Term Review process and should be examinedduring supervision missions to chart progress.

Re-examine environment and strategy. Thescenario team should review for major prob-lems, adjust objectives based on observedresults, and revise time tables and priorities.

The team also should rethink scenario plots inlight of new developments (ideal at Mid-TermReview) and adjust them so that they corre-spond to the most recent information.

Common Pitfalls in Developing Scenarios

Despite its usefulness, scenario analysis is not apanacea. Scenarios are a process, not a product.To become truly effective, they need to be usedand refined over time. This requires sustainedcommitment on the part of high-leveldecisionmakers and a skilled and dedicatedscenario team. It also requires the governmentto carefully develop and track indicators tohelp decisionmakers tie what is going on in theoutside world to the scenarios.

Scenario development is not easy. Theprocess demands significant effort, thought,and creativity of the scenario team. To help thescenario team avoid mistakes made by others,common pitfalls are listed below.

1. Failure to gain the support of key decision-makers. To be credible, scenarios need to beintegrated in the decisionmaking apparatus ofthe organization in which they will be imple-mented. Without real buy-in at the top, thescenarios will not be implemented.

2. Unrealistic goals and expectations. Scenariosdo not produce action plans; they helpdecisionmakers envision what will happen.The methodology is not suited to addressingspecific tactical issues. It is meant to provide abroad view of the uncertainties facing anintervention. Strategic decisions flow from thisunderstanding, but they are not a direct prod-uct of the exercise.

3. Failure to develop a clear map of the future withmonitorable indicators. It is essential that the teamdevelop clear, monitorable milestones of progresstoward the various scenarios. These milestonesshould have a direct relationship to the goals andplanned outcomes of the intervention.

4. Scenarios that are not credible. Scenario work-shops are not brainstorming sessions. Scenariosmust be based on solid quantitative as well asqualitative projections if they are to be credible to

81


those implementing them. Because scenarios donot assign probabilities or project against currenttrends, it is important to make sure that they arebased on strong research.

5. Scenarios that are not tied to the planningprocess. The indicators and thinking in sce-narios must be directly built into the way thatan intervention is planned. Scenario indicatorsshould be closely monitored and associatedwith explicit changes in strategy, including exitstrategies. Similarly, scenarios should berelated to the client’s and the Bank’s budgetand policy cycles.

6. Not enough time to carry out the scenarioprocess. The process requires discipline andattention. It can be divided in two phases:scenario building and relating scenarios to strategy.Although these two phases may take place atseparate times to conform with the Bank’sproject cycle and the client’s policy cycle, thescenario team should have enough time tothink through the logic of the scenarios andensure that they are properly researched.

7. Inappropriate time frame and scope. Scenariosthat focus on current crises and existing prob-lem areas rather than looking at the interactionof broader forces do not generate the kind of newthinking necessary to jump-start an agenda.

8. Mistaking projections for scenarios. Projec-tions are based on past data and often posit acontinuation, or slight variation, of currenttrends. Projections are not well suited to dealingwith the potential for significant discontinuitiesbrought on by external events. Scenarios aredesigned to highlight “what if?” situations fordecisionmakers. Scenarios should be builtaround the forces that shape society. Trends aresymptoms of these forces rather than theircause. These unexpected changes cannot bepredicted with a study of current trends.

9. Failure to tell a dynamic, internally consistentstory. Scenarios should be movies, not still

frames. Each scenario should be a smoothnarrative that makes intuitive sense to thereader. The main aspects of the future shouldbe internally consistent; the outcomes postu-lated for the two key uncertainties should beable to coexist; and the actions of stakeholdersshould be compatible with their interests.

10. Lack of diversity of inputs. If the scenarioteam members are of homogeneous educa-tional backgrounds and institutional affilia-tions, they will be much less likely to come upwith innovative solutions. To build successfulscenarios, the participation of a diverse groupof people is essential.

Conclusion

If scenario development is carried out withcommitment and purpose, it can be a powerfultool for improving development interventions.Scenarios are useful tools for task managersoperating in environments in which long-termthinking, flexibility, and the inclusion of stake-holders in decisionmaking are needed. Sce-narios bring insight to the potential futures inwhich an intervention will operate, what it willneed to do to succeed in each future, and whatneeds to be done to make that success possible.By addressing rather than minimizing uncer-tainty, scenarios spur innovative and robustsolutions. Because they are developed with ateam of knowledgeable stakeholders, they arean effective way to gain buy-in for strategies.Furthermore, the research and discussion thataccompanies scenario development promotecollaborative learning by both World Bankstaff and the client. Ideally, all members ofthe team walk away with a better under-standing of the interactions among thecountry’s social, economic, political, andtechnological forces.

Note

1. The steps that follow owe a significant debt toP. Schwartz, The Art of the Long View.

82

Jonathan N. Maack

Appendix 1. Preparation

The first step in the scenario analysis process isfor the task manager to define the interventionwith a counterpart in the client government.The task manager decides how scenario analy-sis will be used in the project and works out ageneral plan to incorporate it in the projectcycle. She or he then assembles the scenarioteam in consultation with the client govern-ment and local and international NGOs. Oncethe team is assembled, the task managerconducts a series of one-on-one, preliminaryinterviews to determine the team members’points of view regarding the intervention.These interviews can help the guide theorganizers of the exercise in assemblinguseful background material before the firstworkshop.

Scenario Team

In the broadest sense, the scenario team is thegroup of leaders and experts who gather tocreate four or five distinct narratives about thefuture over the course of several workshops.Each of the members gives his or her time,energy, leadership experience, and technicalexpertise to the group for the duration of theworkshops. Team members should be chosenbased on their ability to represent distinctviewpoints on the issue being discussed, be ittechnical or political. Ideally, all also will bechampions of the scenario process and generatesupport for its ideas among their colleaguesand communities.

There is also a smaller group within thescenario team that plays an active role before,between, and after the series of workshopsduring which the scenarios are created. Thissmaller group, known as the core scenario team,should be led by key decisionmakers at theBank and in the country. The finished scenariosneed to be authored by, or in collaborationwith, the people who will be asked to lead theimplementation of the intervention. Othermembers of this smaller core team will conduct

preliminary interviews, analyze the countrysituation, and ground-truth the assumptions ineach scenario.

Assembling the Scenario Team:A Preliminary Checklist

Although it may not be possible to includemultiple representatives from each group listedbelow, the best scenario teams are diverse.Below is a list of individuals likely be part of ascenario team:

1. Decisionmakers. The task manager and asenior representative from the borrower gov-ernment should jointly lead the process andwrite the final scenarios.

2. People with a thorough knowledge of theWorld Bank and its role in the country. A represen-tative from the Country Office or the taskmanager can fill this role.

3. People with an understanding of the borrowergovernment and its role in the sector/field (whenappropriate). This includes senior staff at theappropriate ministry and officials who areinterested in the scenario process and are activein the field.

4. Experts and specialists. This includesBank, engineers/technical specialists, localacademicians, and other relevant expertsfrom other donor agencies or internationalNGOs.

5. Line ministry staff and/or implementingNGOs. These individuals will carry out theoperation if it is an investment loan. In astrategy context, the net can be cast morebroadly, but there is nonetheless a clear needfor practitioners to give input on the realities ofimplementation.

6. Leaders from the private sector and from tradeunions. In an investment loan, these representa-tives should come from the industry or fieldmost closely related to the project. In a strategycontext, a diverse group of dynamic business-persons can be considered. It is important that

83


both local and international interests be repre-sented when they will be equally affected bythe intervention.

7. Leaders from NGOs, activist groups, religiousorganizations, and beneficiary organizations.

8. Politicians. Members of all relevant politi-cal parties should be included in strategydiscussions. In Colombia, politicians includedguerillas and revolutionaries.

Preliminary Interviews

Preliminary interviews engender comfort withthe project in the participants, give the team asense of individual styles, and help identifypotentially explosive issues. It is essential thatall participants in the scenario process beinterviewed. This includes Bank staff andmembers of the client government. The keyoutcome of these interviews will be a betterunderstanding of the mental models (theoutlook and expectations) of each member ofthe scenario team.

Most interviews will take between one andtwo hours to conduct. Interviews can be con-ducted by Bank staff, members of the clientgovernment, or staff from an external agency orNGO that is running the process. The idealinterviewer would be a locally based indi-vidual or team with experience in facilitationand an understanding of the policies of boththe client government and the Bank. Interview-ers should be fluent in the language of thoseinterviewed (interviews should not be con-ducted through an interpreter unless absolutelynecessary) and be fully conversant with boththe details of the project and the scenarioprocess. Interviewers must establish trust andcredibility with participants from the begin-ning. This can be accomplished by clearlyexplaining the goals of the process and high-lighting the fact that the data collected willremain confidential.

Interview Questions

Interviews are meant to be conversations ratherthan formal surveys. The interviewer shouldparticipate in a reactive rather than active role,feeding back responses and asking questions to

clarify what has been said. The challenge in thissort of interviewing is to remain unobtrusiveand affect the interviewee’s train of thought aslittle as possible. The following list of open-ended questions should help interviewers tomove the conversation without directing it. Allquestions should be phrased so that they applyprimarily to the organizations/situations withwhich the interviewee has the greatest interac-tion/association.

Clairvoyant: If I could answer three questionsabout the future for you, what would they be?This question helps interviewers identify thegreatest sources of uncertainty while encour-aging the interviewee to prioritize theirconcerns.

Good scenario: Imagine that the future werevery good, happening exactly as you would wish.How would you answer the three questions that youasked the clairvoyant? What would have to happento cause this future? This question encouragesthe interviewee to revisit the three uncertaintieshe/she introduced in his or her first questionand resolve them while creating a story inwhich all the elements interact.

Bad scenario: Imagine that the future developedalong the worst possible lines. How would youanswer the three questions you asked the clairvoy-ant in this case? What would have to happen tocreate this future?

Lessons from the past: What does (the country/the government, your agency/NGO/the WorldBank, the agency that provides you services/thesector or field) need to forget? What does it need toremember?

Important decisions ahead: What are one ortwo critical strategic decisions on the immediatehorizon for your organization/group?

Constraints in the system: What are theobstacles to this process? To the “good scenario”?This question should elicit strong responsesand will be closely tied to the organization theinterviewee knows best.

Epitaph: What would you like to be rememberedfor having accomplished? This very open-endedquestion allows for identification of the role ofthe organization and individual’s goals andcommitment.

Closure: What should I have asked you that Ididn’t?

84

Jonathan N. Maack

Visual Clustering to Analyze Interview Results

Immediately after the interview, interviewersshould sit down for at least 30 minutes to gothrough their notes to identify importantobservations to be processed further. A simpleguideline for identifying key points is: “Is theview expressed relevant and significant in thecontext of the organization’s relationship withbeneficiaries and the external environment?”Issues that are related to internal processes ofthe organization should be included onlyinsofar as they affect the way that the inter-viewee or organization will react to the outsideworld. Interviews normally yield from 40 to 60statements. Each statement should be reducedto a single line.

In the next stage of analysis, the interviewercombines the statements from all of the inter-views to create logical cause-and-effect chainsof reasoning. At first, statements will seem likerandom, unconnected thoughts, but as inter-viewers read through them, patterns andconnections should emerge. Clustering, orgrouping these statements, serves three pur-poses: (1) it forces the analysts to pay attentionto the conceptual meaning of what has beensaid; (2) it organizes the thoughts of a diversegroup into more coherent stories; (3) it is aniterative process that promotes serial appraisalof the outlooks of each of the people inter-viewed. This technique is best suited to clustercomments from a set of 5 to 10 interviews. It isimportant to remember that with only 9 inter-views, the core scenario team could be dealingwith as many as 450 statements.

However, up to 25 persons can participate ina scenario exercise. The interviewer can simplytake the statements from several related inter-views and cluster them separately. Say, forexample, there are 20 interviews, and 5interviewees are from the Bank, 6 are fromNGOs, 4 are from the client government, andthe other 5 have no particular affiliation. Thestatements of NGO representatives could beclustered separately, as could those of the otherrelated groups. Clustering enables the inter-viewer to work with a manageable number ofstatements. Each related group of intervieweeswill then yield a smaller number of “clustered”statements that synthesize the thinking of the

group. The final stage consists of using the“clustered” statements from each related groupof interviews, which should not total more than15 statements per group of interviews, as thebasis for a last clustering to form a final, overallmap of interview responses.

Visual Clustering Process

The core scenario team should write the state-ments from the interviews on adhesivenotecards and place them on a display surfaceor a wall. Statements should be written inlarge, heavy characters so that they can be readfrom a distance and text should be limited to afew (at most 8) words. As team members readthe statements, they should be able to notepatterns emerging. The idea is to organize theseparate comments into higher-level conceptsthat can be related to one another.

1. When clusters are finalized, team mem-bers should rephrase each one into a one-wordstatements that sum up the logic of the variousstatements brought together to form it. Forexample, a key internal constraint to educa-tional planning is the lack of efficient disburse-ment mechanisms, which affects both teacherpay and procurement.

2. Statements that do not fit naturally intoexisting categories should be put aside andseparate categories created for them later.

3. As the process progresses, team mem-bers begin to arrange the clusters hierarchi-cally according to higher level concepts. Forexample, it may be possible to sum up 15statements about disbursement in a cluster,which should then be organized under alarger cluster focusing on, say, internalbudgetary issues.

At the end of the process, there should befive to six larger concepts and enough clustersto ensure that no cluster holds more than 15 orso statements. Statements should flow fromsimple clusters to larger ideas.

Assembling Background Information

The clustered results of the preliminary inter-view should give the core scenario team an

85


idea of the expertise of the group and theinformation that team members are likely toneed to make informed judgments whenbuilding scenarios. If possible, the core teamshould draw together basic information onrelevant issues and present it to the group asa packet. This background can help thegroup start off with the same basic referencepoints.

The collection of relevant informationregarding these forces should focus on thecountry itself, its neighbors, trading partners,and (when appropriate) regional trends. Thegeneral areas for preliminary data gatheringare:

• Predictable, relevant forces. This informationmight include short-term economic forecasts,relevant demographic data, household type

and composition, and population growthrates.

• Unpredictable, relevant forces. Unpredictableforces such as the price of oil, internationalinvestment flows, aid levels, and the sustain-ability of debt can significantly affect anintervention. Research into these topics ismore involved. Industry and academicexperts could be consulted, and historicaltrends affecting these forces could beassembled to aid comparisons with thecurrent situation.

• Sociopolitical information. A focused analysis ofthe historical and political situation in thecountry as it relates to the intervention shouldbe assembled for the benefit of outsiders. Itcould become the focus of a brief discussionduring the workshop to enable people fromthe affected areas to add detail.

86

Jonathan N. Maack

Appendix 2. A Roadmap to the Scenario Analysis Process

Preparation

Build Scenarios

Reiterate and perfect

Use Scenarios to Plan Strategy

Implement Strategy

Investigate and Write Scenarios

· Assemble group that is representative and diverse· Conduct interviews to determine mental models of participants· Carry out preliminary research based on key issues revealed· Schedule and arrange workshops

· Determine decision focus (1 hr)· Identify key factors and driving forces (3-4 hrs)· Prioritize forces using scenario matrix (2 hrs)· Write basic scenario plots (1 hr)

· Rethink scenario logics (1-2 hrs)· Flesh out scenarios in plenary session (2 hrs)· Break out in groups to finish scenarios (4 hrs)· Present scenarios to groups and wrap up (1 hr)

· Conduct quantitative and qualitative research (3 wks)· Write full-length drafts of each scenario (2-3 wks)

· Correct, revise, and amend draft scenarios in plenary sessions (1 hr)· Explore strategic implications of each individual scenario (3 hrs)· Develop monitorable indicators for project success in each scenario (2 hrs)· Highlight “trigger” events that will signal environment is moving toward one of the scenarios (2 hrs)

· Simplify scenarios and prepare them for dissemination· Gain support from public and those who will implement scenarios (ongoing)

· Build accountability and incentive systems· Implement changes on the ground (ongoing)

· Monitor progress and test results (ongoing)· Reexamine environment and strategy (ongoing)

Workshop I

Disseminate scenarios Set up implementation strategy

Workshop II

Workshop III

87


References

Bonnet, T. W., and R. L. Olson. 1998. “How ScenariosEnrich Public Policy Decisions.” In Learning fromthe Future: Competitive Foresight Scenarios, editedby L. Fahey and R. M. Randall. New York: JohnWiley & Sons. 308-24.

Butcher, J. 1999. “Advance Interviewing for ScenarioAnalysis.” San Francisco: Global Business Net-work. (Unpublished presentation.)

Fahey, L., and R. M. Randall, eds. 1998. Learning fromthe Future: Competitive Foresight Scenarios. NewYork: John Wiley & Sons.

_____. 1998. “What Is Scenario Learning?” InLearning from the Future: Competitive ForesightScenarios. 3-21.

Fahey, L. and R. M. Randall. 1998. “IntegratingStrategy and Scenarios.” In Learning from theFuture: Competitive Foresight Scenarios. 22-38.

Georgantzas, N. C., and W. Acar. 1995. Scenario-Driven Planning: Learning to Manage StrategicUncertainty. Westport, Conn.: Quorum Books.

Godet, M., and F. Roubelat. 1985. “Creating theFuture: The Use and Misuse of Scenarios.” Long-Range Planning 29 (2):164-71.

Kahane, A. 1999. “Destino Colombia.” Deeper News 9(1). Berkeley, Cal.: Global Business Network.

_____. 1998. “The Mont Fleur Scenarios.” DeeperNews 7 (1). Berkeley, Cal.: Global BusinessNetwork.

Kaplan, R. S., and D. P. Norton. 1996. The BalancedScorecard: Translating Strategy into Action. Cam-bridge, Ma.: Harvard Business School Press.

Noonan, P. S., and M. S. Tenaglia. 1998. “HowInformation Technology Helps ScenariosAdvance from Consensus to Decisiveness.” InLearning from the Future: Competitive ForesightScenarios. 384-401.

Ringland, G. 1998. Scenario Planning: Managing for theFuture. New York: John Wiley & Sons.

Schoemaker, P. J. H. 1998. “Twenty Common Pitfallsin Scenario Planning.” In Learning from the Future:Competitive Foresight Scenarios.422-32.

P. Schwartz. 2000. “The Official Future, Self-Delusionand the Value of Scenarios.” Financial Times.“Mastering Risk,” Part Two. May 2.

_____. 1996. The Art of the Long View: Planning for theFuture in an Uncertain World. New York:Doubleday.

Schwartz, P., and J. A. Ogilvy. 1998. “Plotting YourScenarios.” In Learning from the Future: CompetitiveForesight Scenarios. 57-80.

Van der Heiden, K. 1998. “Setting the Agenda.” SanFrancisco: Global Business Network. (Unpub-lished.)

Wilson, I. 1998. “Mental Maps of the Future: AnIntuitive Logics Approach to Scenarios.” InLearning from the Future: Competitive ForesightScenarios. 81-108.

5. Scenario Analysis: A Tool for Task Managers · 5. Scenario Analysis: A Tool for Task Managers...

Documents

Transcript of 5. Scenario Analysis: A Tool for Task Managers · 5. Scenario Analysis: A Tool for Task Managers...