## htnea2021010-s001

## Source details

**Canonical URL:** [htnea2021010-s001](https://www.imf.org/-/media/files/publications/analytical-notes/2021/english/htnea2021010-s001.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/analytical-notes/2021/english/htnea2021010-s001.pdf.md)
- [Structured JSON version](/-/media/files/publications/analytical-notes/2021/english/htnea2021010-s001.pdf.json)

---

### One-day workshop segment: stocktake, classification, interactions, plenary
- Schedule and main activities:
  - 12.20 pm Lunch
  - 13:00 pm Stock Take of Trends/Tensions, Drivers, Variables and Links to Draft Surveillance Priorities
  - 13:30 pm Groups Work on Classification
    - Split landscape headings into 3 groups.
    - Each group arranges trends/tensions subcomponents and or drivers along axes for (i) degree of uncertainty and (ii) relevance to CSR surveillance priorities.
    - Each group chooses 3-4 most relevant (with varying degrees of uncertainty).
  - 14:30 pm New Groups Work on Interactions
    - Mix participants into 3 new groups.
    - Tasks:
      - (a) Make a prediction on the interactions using 1 of the less uncertain trends/tensions/drivers from each previous group.
      - (b) Create 2 stories showing possible differences depending on the outcome of at least one of the more uncertain trends/tensions/drivers from a group and 1-2 others.
    - Short break coffee (provided) and comfort break.
  - 15:45 pm Plenary and Feedback (including project supervisors joining at this stage)
    - 45-60 minutes including clarification questions and comments from supervisors.
    - Recap of stocktaking exercise, including which CSR headings are more or less important and if any should be dropped/downplayed/replaced.
    - Introduction of underlying drivers and variables.
    - Group nominees present:
      - Classification of most relevant and most uncertain trends/tensions/drivers.
      - Illustration of interactions.
- Post-workshop decisions to inform interim note and full draft of the CSR:
  - Decide which landscape headings to emphasize.
  - Decide whether deeper drivers can be accommodated within those headings or require separate articulation (e.g., a box).
  - Decide which subcomponents matter more.
  - Decide which developments should be cast as likely vs illustrative uncertainties.
  - Include interactions between trends/tensions/drivers in the scoping note but not the full CSR.
  - Prioritized uncertainties will feed into scenarios.

### Online Annex 4: 2-Day Deductive Method Scenario Building Agenda (goal: outline 2-4 scenarios by end of Day 1)
- Day 1 (timed agenda highlights):
  - 8:15 to 8:30 am Breakfast
  - 8:45 to 9:15 am First Deputy Managing Director opening remarks
  - 9:15 to 10:30 am Plenary session I: (I) Introduction to scenario development by Professor Rafael Ramirez; (II) Walk-through of the two-day program by Professor Ramirez; (III) Presentations by external specialists
  - 10:30 to 10:45 am Coffee break
  - 10:45 to 11:45 am Group work 1: 4 Groups to identify a broad set of factors affecting the IMF’s contextual environment; three-minute presentations by each participant at the beginning
  - 12:00 to 12:45 pm Group work 2: Same groups to pinpoint the two most important factors
  - 12:45 to 1:45 pm Lunch
  - 1:45 to 2:15 pm Group work 3: Two sets of two groups compare the most important factors
  - 2:15 to 2:30 pm Group work 4: Within groups discussion about what they have learned from the other group in the respective pair
  - 2:30 to 2:45 pm Break
  - 2:45 to 3:45 pm Group work 5: Same group pairs decide on the 2 most important factors and develop the contours of one or two sufficiently different scenarios; begin thinking about IMF interactions with existing or emerging new players
  - 3:45 to 4:15 pm Coffee break
  - 4:15 to 6:00 pm Plenary session II: Professor Ramirez to facilitate a plenary to agree on the broad outlines of the scenarios
- Day 2 (goal: final set of scenarios by end of day):
  - 8:15 to 8:30 am Breakfast and check-in
  - 8:45 to 9:15 am Brief Plenary session III (Prof. Ramirez to recap)
  - 9:15 to 10:15 am Group work 6: 2 to 4 groups; each group works on (i) one scenario chosen the previous day and (ii) contextual and transactional environment in each scenario
  - 10:30 to 11:30 am Group work 7: Ensure comparability/consistency across scenarios
  - 11:45 am to 12:30 pm Group work 8: Refine and write scenario narrative, including "sub-stories" (for instance, related to geopolitics)
  - 1:30 to 3:30 pm Group work 5: Enrich storyline and prepare a brief presentation
  - 4:00 to 5:15 pm Plenary session IV: Deputy Directors from all departments attend presentation of the scenarios
  - 5:15 to 6:00 pm Plenary session V: Closing remarks
- Notes:
  - "Selection of axes prompted far more debate than anticipated, so this target was only achieved the next morning."
  - Follow-up invitations may be used to seek feedback on draft scenarios a few weeks later.

### Online Annex 5: 2-Day Inductive Method Scenario Building Agenda
- Day 1 (high-level flow):
  - 8:30 am Welcome, introductions, and objectives (Mainly to stress-test surveillance priorities; also to demonstrate uncertainties in the surveillance landscape and to facilitate engagement with stakeholders).
  - 8:50 am Imagining the future: Warm-up in 4 groups of about 4, followed by short plenary.
  - 9:10 am Contextual environment: Group work (60 mins) to identify most important factors driving macro-financial surveillance landscape in 2030; groups agree two most important trends/uncertainties.
  - 10:45 am Short Presentation on inductive method using factor/uncertainty cards.
  - 10:55 am Mini scenarios set I (45 minutes): 4 groups of about 4 people; individuals choose a card printed before the workshop and explain why; sketch scenarios using stylized systems map; give placeholder name and 1-2 sentence summary; nominate an observer.
  - 11:40 am Plenary presentation of drafts from round I. Facilitator feedback and pointers.
  - 12:10 pm Observers give within-group feedback on group dynamics. Appoint process lead for next round. Choose cards for next round.
  - 12:30 pm Lunch
  - 1:15 pm Mini scenarios sets II and III: iterative card-based rounds using dice, new groups, and facilitator feedback to build contrasting scenarios; plenary presentations and facilitator-led vote on preferred scenarios at 4:30 pm
  - 5:15 pm Close followed by debrief and proposals within smaller group on which scenarios to keep/merge
- Notes on compressed application:
  - For the International Monetary System scenarios, agenda compressed into two half days with adaptations: (i) early-round issues reproduced on slide instead of brainstorming, (ii) only two rounds of mini scenario building (done in 5 groups), (iii) less time to enrich scenarios; core drafters had to fill in more gaps.
- Day 2 (inductive) highlights:
  - 8:00 am Smaller group including facilitators meets to decide which scenarios to keep/merge.
  - 9:00 am Merging and Enriching scenarios I (45 minutes): deepen scenarios and test causal logic; optionally use economic models to check robustness.
  - 10:45 am Merging and enriching scenarios II: add regional/country variation, timeline, winners and losers.
  - 11:45 am Transactional environment: behavior/preference of central banks, finance ministries, other authorities, economic actors; pressures; new actors.
  - 1:15 pm Plenary preparation and presentation; 1:45 pm Plenary: each group presents scenarios and solicits feedback.
  - 3:30 pm Interdepartmental Surveillance Contacts feedback: ISC evaluate scenarios on plausibility, internal consistency, novelty/surprise, and challenge for surveillance priorities.
  - 5:30 pm Close

### Online Annex 6: Scenario Building Application Workshops (short-form applications and stress testing)
- ANNEX 6.1: Apply scenarios in under 75 minutes with almost 75 people (Communications department off-site retreat)
  - Circulate scenarios document in advance.
  - 10-minute presentation and 3-minute scenarios video.
  - Divide participants into nine groups; each focuses on one question for one scenario (three questions and three scenarios).
  - Groups: 5 minutes internal discussion, 25 minutes on assigned question.
  - 25-minute plenary: each group conveys answer in about 5 words to form a 3X3 matrix of answers; selective elaboration and reflection.
- ANNEX 6.2: Stress testing the Fund’s business model (Heads of Department workshop, Innovation Lab)
  - Pre-workshop: department heads read three scenarios and prepare answers to questions associated with one scenario.
  - 9:00 am–9:10 am Opening by the Managing Director
  - 9:10 am–9:20 am Preview by facilitator, followed by 3-minute scenarios video
  - Three rounds of group work with rotating moderators and scenario teams accompanying each group; facilitator guides specificity
  - 11:35 am–11:55 am Coffee Break where participants vote to prioritize implications anonymously using a secure online survey method.
  - 11:55 am–12:30 pm Managing Director facilitates discussion on top 6 implications: identify decisions Management can take today to shape the IMF’s strategy for the next 25 years.
- Key procedural elements across agendas:
  - Repeated use of small groups and plenary sessions to iterate from broad factors to final scenarios.
  - Use of tools: factor/uncertainty cards, dice, stylized systems maps, presentation templates, and optionally economic models.
  - Explicit steps for testing plausibility, internal consistency, novelty, and relevance for surveillance priorities.
  - Inclusion of senior management and interdepartmental review (e.g., Deputy Directors, ISC) to validate and challenge scenarios.

### ANNEX 6.3: Stress-testing draft proposals using scenarios — workshop agenda and process
- Morning scenario rounds (each scenario treated similarly):
  - 8:15 am Welcome breakfast, introductions, and objectives.
  - 8:40-10:00 am Scenario A: Split into four groups of about 3-4, one group for each of the four surveillance priorities. One person is process lead and a second is a minute taker for this scenario.
    - 10 mins Scenario familiarity exercise: reread scenario and preparatory notes.
    - 45 mins Discuss:
      - (i) How can your priority be illustrated in this scenario? Be specific about policies.
      - (ii) What goes wrong for our members and/or the global economy if these priorities are not addressed?
    - 25 mins Plenary session: Presenters summarize answers from group work. Feedback. Record on laptop/screen.
  - 10:10-11:30 am Scenario B (as before, followed by short break, new minute taker and process lead)
  - 11:40 am-1:00 pm Scenario C (except some people not specifically assigned to a priority switch groups)
  - 1:00 pm Lunch
- Afternoon pre-mortem and ISC integration:
  - 2:00 pm Plenary:
    - (1) Recap of morning, cross scenario comparison, especially what goes wrong (fill in table with priorities in rows and scenarios in columns).
    - (2) Introduction to pre-mortem. Groups (5 each) split into 3 scenarios with representation across the priorities.
  - 2:15-4:00 pm Pre-mortem: Group work structured into three 20-minute blocks:
    - 20 minutes of failure: What could the IEO say went so wrong in this scenario? Each member writes ideas in 7 minutes; share for 10 minutes.
    - 20 minutes of criticism: What would the IEO criticize about CSR surveillance priorities and what did the IMF get wrong? Each member writes in 7 minutes; share for 10 minutes.
    - 20 minutes of reflective solving: Are surveillance priorities missing something? If adding a 5th priority what would it be? Would you alter existing priorities or subcomponents?
    - 5 minutes of voting: Group votes for top 2-3 failure/criticism combinations and 1 (or maybe 2) 5th priority proposals. Idea originators share with plenary.
  - Plenary 35 minutes: For each scenario: scenario recap by facilitator, group feedback by nominees (5 minutes each). Recorded on table on screen. Free-flowing general discussion.
  - 4:00 pm Tea including Interdepartmental Surveillance Contacts (ISC)
  - 4:10-5:10 pm ISC Session, plenary introduction (CSR teams collect thoughts for next round)
    - ISC/CSR groups session steps include short presentations, listing failures/criticisms, voting on surveillance priority adjustments, and selection of top proposals.
  - 5:10 pm Plenary: Each table reports back on 2 changes to SPs to the other groups (10 minutes). Closing comments / reactions on commonalities across scenarios (10 minutes).
  - 5:30 pm Close

### Dice-based deals and success rates (Policy gaming mechanics)
- Dice-based deals:
  - In place of making an argument, each party to the deal rolls the 2 dice to determine whether or not the deal yielded the intended benefit for that player.
  - A sum of 5 or higher is needed for success.
  - Statement on success probability: Deals benefit from a disproportionate chance of being successful (each signatory must roll a 5 or higher to receive the intended impact of the deal, with an 83% chance of success).
- Time horizon for a turn:
  - A game turn typically occurs over a period of about a year (i.e., giving time for trade tariffs to take effect), though length may be variable based on events.
- Exogenous shocks and uncertainty:
  - “It’s a world of heightened uncertainty, so there might be some unexpected surprises.”
- Argument scope and adjudication:
  - Players can argue about abilities, existence of people/places/events, weather/disease/public opinion, expected effectiveness of actions (e.g., fiscal multipliers).
  - Constraints and clarifications include:
    - Players cannot physically block a policy by argument but can argue about likelihood of implementation.
    - The only criterion for judgement is the likelihood of an event/action being successful.
    - Players limited to one action a turn.
    - Contradictory arguments across turns are discouraged as logically inconsistent.
    - Repeated failed attempts usually reduce credibility for future attempts unless circumstances change.
- Measures of success in the game:
  - Emphasis is on process and learning from action-reaction sequences.
  - Meeting one’s objectives (one pre-assigned, one written by player) is a barometer of success; progress checked mid-game and at the end.
  - Number of successful proposed actions/negotiations is another measure.

### Policy Game: Matrix Game mechanics and examples of actor actions
- Matrix Game mechanics (Pros and Cons System):
  - Active player states an action, expected impact, and reasons (Pros) for success.
  - Other players add objective reasons why it might or might not be successful (Additional Pros + Cons).
  - Two adjudication methods for dice thresholds described; Method 1 maps group-offered probability (e.g., 70 percent) to required roll; Method 2 adjusts required roll by pros/cons counts.
  - Dice are rolled to determine success; outcomes inform narrative (marginal vs resounding success).
  - A turn ends when all players have attempted an argument.
- Negotiation and deals:
  - After initial turns or day, deals can be negotiated; at start of second session deals must be announced and agreed verbally with intended specific impacts stated.
- Examples of Actions by actor type:
  - Government Actors: make deals with vaccine producers; diplomatic/military agreements; fiscal changes; monetary policy adjustments; structural reforms; trade measures; financing requests.
  - Multilaterals: organizing vaccine production/distribution; surveillance; introducing new lending facilities (e.g., pandemic loan facility); catalyzing development partners; capacity development; setting institutional views/global standards; changing strategic focus; convening stakeholders.
  - Private Companies: allocating vaccines; mergers and acquisitions; R&D; lobbying; hiring; publicity campaigns; industrial espionage/bribery; venture capital/IPOs; government loans and partnerships.
- Note: Actions should leverage country- or actor-specific factors.

### Facilitation guidance (Online Annex 7)
- Purpose and role of facilitators:
  - External facilitators used while developing in-house capacity; examples include Oxford Scenarios Program, foresight conferences, and facilitation training tailored to IMF culture.
  - Early facilitator tasks: identify purpose and user; interrogate objectives; calibrate resources and division of labor; clarify facilitation is not outsourcing.
- Tailoring workshops:
  - Account for human resources and deadlines; ensure drafters and good feedback providers attend; write scenarios to challenge participants.
  - Decide acceptable risk level to foster insights vs meeting objectives.
  - Facilitators produce scenario-planning timeline and individual workshop agendas; project supervisor must agree and communicate them.
  - For trends workshops: define deliverables and allocate time for research, brainstorming, and prioritizing.
  - For scenario planning: fit steps into available time.
  - For implications workshops: craft questions to elicit candor and specificity.
  - For policy gaming: identify shocks, roles, players, and rules.
- Facilitator functions and logistics:
  - Role described as a “midwife” (Wilkinson, 2017, page 5).
  - Logistics: rooms with good light/acoustics, refreshments, stationery, pre-populated templates.
- General facilitation principles:
  - (1) suggest turning off or handing in mobile phones;
  - (2) limit comment length to allow all to speak without interruption;
  - (3) emphasize that no idea is a bad idea, though ideas can be challenged;
  - (4) designate workshops as a safe space for difficult conversations.
- Foresight facilitation advice:
  - During trends workshops, surface assumptions by asking “Why” more than once.
  - Facilitators may act as devil’s advocates and should challenge implicit extrapolations.
  - During scenario-building, push to stretch plausibility, avoid scenarios being too similar, and help evaluate scenarios as a set.
  - Facilitators help project leaders make judgement calls like scenario selection and can provide assertive advice if workshops hit an impasse.
  - For implications workshops, encourage precise and frank answers; for policy-gaming, encourage precise pro and con arguments.

### Online Annex 9: Options for Foresight Engagement — comparison of methods (benefits, risks, resource costs)
- Full Scale, Wide Angle Scenario Planning
  - Positives:
    - Meets periodic need to inform strategic direction.
    - Offers respite from daily activities.
    - Enhanced familiarity and ownership by user (especially top-level).
    - Experience of building scenarios is challenging and insightful, and ultimately more influential on subsequent stages.
  - Negatives/Risks:
    - Without existing demand or operational directive, value can be harder to demonstrate in advance.
    - Facilitation by junior foresight team or outside experts have drawbacks; lack of senior in-house staff may be limiting.
  - Resource Costs:
    - 3-4 days direct involvement by user possibly with top-level and senior staff though can be spread over a long period (e.g., months or years).
    - Considerable preparation from foresight team, especially for top-level events.
- Targeted Scenario Planning
  - Positives:
    - More potential for a direct output and actionable conclusions.
    - Synergies between staff work needed for direct output and for scenario planning.
  - Negatives/Risks:
    - Can be insufficiently aware of broader factors without recent wide-angle exercise or in-built research.
    - Topic chosen, if operational, can be “loaded” and participants may hold positions rather than an open mind.
  - Resource Costs:
    - 4-8 hours direct involvement by user.
    - Low/no marginal cost on staff preparatory work on trends and uncertainties.
- Engagement with Prepared Scenarios (premortems, gamification, targeted questions)
  - Positives:
    - Can leverage existing materials if the existing scenario set is still relevant.
  - Negatives/Risks:
    - Fewer topics can be addressed; less ownership and experiential learning.
  - Resource Costs:
    - 2-4 hours for user.
- Policy-gaming (strategic simulation exercises with roleplay)
  - Positives:
    - Experiential learning by taking on a role; helps develop contingency planning; facilitates team building.
  - Negatives/Risks:
    - Can amplify biases/priors; enjoyment can cloud learning; gamification may appear dismissive of real-world stakes.
  - Resource Costs:
    - ½ day for user plus trial runs; moderately high prep needs; careful adjudication for top-level events.
- “Unknown Unknowns” & Breaking Trends Workshop
  - Positives:
    - Limited prep possible; less structured format may encourage candid sharing.
  - Negatives/Risks:
    - Less likely to unearth “ah-ha” findings; hard to produce meaningful output in short time.
  - Resource Costs:
    - 2-4 hours with user.
- Pre-mortem
  - Positives:
    - Useful for highly targeted questions; resource efficient; intuitive.
  - Negatives/Risks:
    - If participants do not find the failure scenario plausible, it can prevent useful discussions or dissuade action.
  - Resource Costs:
    - 2 hours with users.
- Tabletop exercise (TTX) / crisis simulation
  - Positives:
    - Decision-makers play themselves; simulation typically occurs over multiple days with several in-game updates.
  - Negatives/Risks:
    - Highly resource intensive for facilitators and participants; if participants do not find failure scenario plausible, they may not take it seriously.
  - Resource Costs:
    - 12+ hours with users.
- Aspirational scenario / visioning exercises
  - Positives:
    - Can engender buy-in in contexts of inertia or risk aversion.
  - Negatives/Risks:
    - Can reflect powerful actors’ biased agendas; may appear naively optimistic.
  - Resource Costs:
    - 1-2 days with users.
- Horizon scanning for tail risks
  - Positives:
    - Highlights potential emergent risks.
  - Negatives/Risks:
    - Can be viewed as a waste of constrained resources.
  - Resource Costs:
    - Resource needs vary largely.

*Source: htnea2021010-s001 - ANNEX 6.3 STRESS-TESTING DRAFT PROPOSALS USING SCENARIOS*

### 12.20 pm Lunch

### htnea2021010-s001 - 12.20 pm Lunch

### One-day workshop segment: stocktake, classification, interactions, plenary
- 12.20 pm Lunch
- 13:00 pm Stock Take of Trends/Tensions, Drivers, Variables and Links to Draft Surveillance Priorities
- 13:30 pm Groups Work on Classification
  - Split landscape headings into 3 groups.
  - Each group arranges trends/tensions subcomponents and or drivers along axes for (i) degree of uncertainty and (ii) relevance to CSR surveillance priorities.
  - Each group chooses 3-4 most relevant (with varying degrees of uncertainty).
- 14:30 pm New Groups Work on Interactions
  - Mix participants, creating 3 new groups.
  - Tasks:
    - (a) Make a prediction on the interactions using 1 of the less uncertain trends/tensions/drivers from each previous group.
    - (b) Create 2 stories showing possible differences depending on the outcome of at least one of the more uncertain trends/tensions/drivers from a group and 1-2 others.
  - Short break coffee (provided) and comfort break.
- 15:45 pm Plenary and Feedback (including project supervisors joining at this stage)
  - 45-60 minutes including clarification questions and comments from supervisors.
  - Recap of stocktaking exercise, including which CSR headings are more or less important and if any should be dropped/downplayed/replaced.
  - Introduction of underlying drivers and variables.
  - Group nominees present:
    - Classification of most relevant and most uncertain trends/tensions/drivers.
    - Illustration of interactions.

- Post-workshop decisions to inform interim note and full draft of the CSR:
  - Decide which landscape headings to emphasize.
  - Decide whether deeper drivers can be accommodated within those headings or require separate articulation (e.g., a box).
  - Decide which subcomponents matter more.
  - Decide which developments should be cast as likely vs illustrative uncertainties.
  - Include interactions between trends/tensions/drivers in the scoping note but not the full CSR.
  - Prioritized uncertainties will feed into scenarios.

### Online Annex 4: 2-Day Deductive Method Scenario Building Agenda (goal: outline 2-4 scenarios by end of Day 1)
Day 1
- 8:15 to 8:30 am Breakfast
- 8:45 to 9:15 am First Deputy Managing Director opening remarks
- 9:15 to 10:30 am Plenary session I
  - (I) Introduction to scenario development by Professor Rafael Ramirez
  - (II) Walk-through of the two-day program by Professor Ramirez
  - (III) Presentations by external specialists
- 10:30 to 10:45 am Coffee break
- 10:45 to 11:45 am Group work 1
  - 4 Groups to identify a broad set of factors affecting the IMF’s contextual environment.
  - Three-minute presentations by each participant at the beginning.
- 11:45 am to 12:00 pm Coffee break
- 12:00 to 12:45 pm Group work 2
  - Same groups to pinpoint the two most important factors.
- 12:45 to 1:45 pm Lunch
- 1:45 to 2:15 pm Group work 3
  - Two sets of two groups compare the most important factors.
- 2:15 to 2:30 pm Group work 4
  - Within groups discussion about what they have learned from the other group in the respective pair.
- 2:30 to 2:45 pm Break
- 2:45 to 3:45 pm Group work 5
  - Same group pairs decide on the 2 most important factors and develop the contours of one or two sufficiently different scenarios.
  - Some designated participants to start thinking about how the IMF would interact with existing or emerging new players in the global economy (for day 2).
- 3:45 to 4:15 pm Coffee break
- 4:15 to 6:00 pm Plenary session II
  - Professor Ramirez to facilitate a plenary to agree on the broad outlines of the scenarios.

Day 2 (goal: final set of scenarios by end of day)
- 8:15 to 8:30 am Breakfast and check-in
- 8:45 to 9:15 am Brief Plenary session III (Prof. Ramirez to recap)
- 9:15 to 10:15 am Group work 6
  - 2 to 4 groups. Each group working on:
    - (i) one of the scenarios chosen the previous day
    - (ii) contextual and transactional environment in each scenario.
- 10:15 to 10:30 am Coffee break
- 10:30 to 11:30 am Group work 7
  - Ensure comparability/consistency across scenarios.
- 11:30 to 11:45 am Break
- 11:45 am to 12:30 pm Group work 8
  - Refine and write scenario narrative, including "sub-stories" (for instance, related to geopolitics).
- 12:30 to 1:30 pm Lunch
- 1:30 to 3:30 pm Group work 5
  - Enrich storyline and prepare a brief presentation.
- 3:30 to 4:00 pm Coffee break
- 4:00 to 5:15 pm Plenary session IV
  - Deputy Directors from all departments attend presentation of the scenarios.
- 5:15 to 6:00 pm Plenary session V
  - Closing remarks

Notes:
- "Selection of axes prompted far more debate than anticipated, so this target was only achieved the next morning."
- Follow-up invitations may be used to seek feedback on draft scenarios a few weeks later.

### Online Annex 5: 2-Day Inductive Method Scenario Building Agenda
Day 1
- 8:30 am Welcome, introductions, and objectives (Mainly to stress-test surveillance priorities; also to demonstrate uncertainties in the surveillance landscape and to facilitate engagement with stakeholders).
- 8:50 am Imagining the future: Warm-up in 4 groups of about 4, followed by short plenary.
- 9:10 am Contextual environment: Group work (60 mins) to identify most important factors driving macro-financial surveillance landscape in 2030; groups agree two most important trends/uncertainties.
- 10:30 am Short Coffee break
- 10:45 am Short Presentation on inductive method using factor/uncertainty cards.
- 10:55 am Mini scenarios set I (45 minutes)
  - 4 groups of about 4 people.
  - Individuals choose a card printed before the workshop and explain why; use them to make a draft scenario yielding outcomes for at least a couple of macroeconomic variables and/or the headings.
  - Sketched using stylized systems map; give scenario a placeholder name and a 1-2 sentence written summary.
  - Nominate an observer to offer within-group feedback on group dynamics.
- 11:40 am Plenary presentation of drafts from round I. Facilitator feedback and pointers.
- 12:10 pm Observers give within-group feedback on previous round group dynamics. Appoint process lead for next round. Choose cards for next round.
- 12:30 pm Lunch
- 1:15 pm Mini scenarios sets II and III
  - 1:15 pm (60 mins) Mini scenarios set II: Use new cards and dice to build new scenarios; after 40-45 minutes, discuss with facilitator; refine; give placeholder name and 1-2 sentence summary; nominate presenter.
  - 2:15 pm Choose new card (or take assigned card) and move to a new group.
  - 2:20 pm (55 mins) Mini scenarios set III: Use new cards and build new batch of scenarios to fill contrasts/gaps; after 40-45 minutes, discuss with facilitator; refine; nominate presenter.
- 3:15 pm Tea
- 3:30 pm Plenary: Present scenarios II and III to other groups and special guest.
- 4:30 pm Facilitator-led discussion/vote on preferred scenarios (plausibility, internal consistency, relevance for surveillance, challenge for SPs, novelty). Facilitators populate basic comparison table during vote.
- 5:15 pm Close followed by debrief and proposals within smaller group on which scenarios to keep/merge

Notes on compressed application:
- For the International Monetary System scenarios, agenda compressed into two half days with adaptations: (i) early-round issues reproduced on slide instead of brainstorming, (ii) only two rounds of mini scenario building (done in 5 groups), (iii) less time to enrich scenarios; core drafters had to fill in more gaps.

Day 2 (inductive)
- 8:00 am Smaller group including facilitators meets to decide which scenarios to keep/merge.
- 8:30 am Opening plenary: reflections and proposed scenario set.
- 9:00 am Merging and Enriching scenarios I (45 minutes): Groups of on average 4 deepen scenarios and test causal logic; optionally use economic models to check robustness.
- 9:45 am (45 minutes) Remedial measures on robustness if needed or enrich scenarios.
- 10:30 am Morning Break
- 10:45 am Merging and enriching scenarios II: further remedial measures or add regional/country variation, timeline, winners and losers.
- 11:30 am Short presentation by facilitator on transactional environment
- 11:45 am Same groups: Transactional environment (behavior/preference of central banks, finance ministries, other authorities, economic actors; pressures; new actors).
- 12:30 pm Lunch
- 1:15 pm Plenary preparation and presentation
  - 1:15 pm: Groups divide so 2-3 people revise presentation template and remainder find a picture.
  - 1:45 pm: Plenary: each group presents scenarios and solicits feedback.
- 2:30 pm Groups revise scenarios based on feedback (cross-scenario comparisons, balance).
- 3:15 pm Tea
- 3:30 pm Interdepartmental Surveillance Contacts feedback
  - 3:35 pm: Nominee for each group presents scenario to senior staff serving as Interdepartmental Surveillance Contacts (ISC).
  - 4:10 pm: ISC discuss 2 scenarios in groups; after 25 minutes staff swap tables to discuss second scenario. ISC asks whether scenario is (i) plausible, (ii) internally consistent, (iii) novel/surprising, (iv) challenging for the surveillance priorities.
  - 5:10 pm Plenary of comments (ISC nominee for each scenario responds).
- 5:30 pm Close

### Online Annex 6: Scenario Building Application Workshops (short-form applications and stress testing)
ANNEX 6.1: Apply scenarios in under 75 minutes with almost 75 people (Communications department off-site retreat)
- Circulate scenarios document in advance.
- 10-minute presentation and 3-minute scenarios video.
- Divide participants into nine groups; each focuses on one question for one scenario (three questions and three scenarios).
- Groups: 5 minutes internal discussion, 25 minutes on assigned question.
- 25-minute plenary: each group conveys answer in about 5 words to form a 3X3 matrix of answers; selective elaboration and reflection.

ANNEX 6.2: Stress testing the Fund’s business model (Heads of Department workshop, Innovation Lab)
- Pre-workshop: department heads read three scenarios and prepare answers to questions associated with one scenario.
- 9:00 am–9:10 am Opening by the Managing Director
- 9:10 am–9:20 am Preview by facilitator, followed by 3-minute scenarios video
- 9:20 am–10:00 am First round—Group work (moderators stay with scenario they prepared; others join new scenario)
  - Moderator presents implications and discussion; groups agree on new moderators.
  - Scenarios team accompanies each group to answer questions and take minutes.
- 10:00 am–10:40 am Second round—Group work (new moderators remain; other members switch to a new scenario). Facilitator guides specificity.
- 10:40 am–10:50 am Coffee Break
- 10:50 am–11:20 am Third round—Group work (all other groups return to original scenario except original moderator). Final amendments; facilitator probes common themes and nuance.
- 11:20 am–11:35 am Facilitator leads discussion on process and explains voting process.
- 11:35 am–11:55 am Coffee Break where participants vote to prioritize implications anonymously using a secure online survey method.
- 11:55 am–12:30 pm Managing Director facilitates discussion on top 6 implications: identify decisions Management can take today to shape the IMF’s strategy for the next 25 years.

Key procedural elements across agendas
- Repeated use of small groups and plenary sessions to iterate from broad factors to final scenarios.
- Use of tools: factor/uncertainty cards, dice, stylized systems maps, presentation templates, and optionally economic models.
- Explicit steps for testing plausibility, internal consistency, novelty, and relevance for surveillance priorities.
- Inclusion of senior management and interdepartmental review (e.g., Deputy Directors, ISC) to validate and challenge scenarios.

*Source: https://www.imf.org/-/media/files/publications/analytical-notes/2021/english/htnea2021010-s001.pdf*

### ANNEX 6.3 STRESS-TESTING DRAFT PROPOSALS USING SCENARIOS

### ANNEX 6.3 STRESS-TESTING DRAFT PROPOSALS USING SCENARIOS

### Workshop agenda and process for Scenario-based stress-testing
- 8:15 am Welcome breakfast, introductions, and objectives.
- 8:40-10:00 am Scenario A: Split into four groups of about 3-4, one group for each of the four surveillance priorities. One person is process lead and a second is a minute taker for this scenario.
  - 10 mins Scenario familiarity exercise: reread scenario and preparatory notes.
  - 45 mins Discuss:
    - (i) How can your priority be illustrated in this scenario? Be specific about policies.
    - (ii) What goes wrong for our members and/or the global economy if these priorities are not addressed?
  - 25 mins Plenary session: Presenters summarize answers from group work. Feedback. Record on laptop/screen.
- 10:10-11:30 am Scenario B (as before, followed by short break, new minute taker and process lead)
- 11:40 am-1:00 pm Scenario C (except some people not specifically assigned to a priority switch groups)
- 1:00 pm Lunch
- 2:00 pm Plenary:
  - (1) Recap of morning, cross scenario comparison, especially what goes wrong (fill in table with priorities in rows and scenarios in columns).
  - (2) Introduction to pre-mortem. Groups (5 each) split into 3 scenarios with representation across the priorities.
- 2:15-4:00 pm Pre-mortem. Group work: Imagine the Independent Evaluation Office is evaluating surveillance looking back over the decade in this scenario in 2030 and concluding it went horribly wrong.
  - 20 minutes of failure:
    - Using the plenary recap or otherwise, answer: What could it say went so wrong in this scenario?
    - Each member writes down as many items as s/he wants in 7 minutes.
    - For 10 minutes, share your ideas with the group and write numbered list on board / screen (or place sticky on board).
  - 20 minutes of criticism:
    - Based on the items above (or otherwise), what would be its criticism of the CSR surveillance priorities?
    - What specifically did the IMF (not the authorities) get wrong and why?
    - Each participant writes down as many as s/he wants in 7 minutes.
    - For 10 minutes, share ideas within the group and match to list from previous bullet on board / screen if applicable.
  - 20 minutes of reflective solving:
    - Are the surveillance priorities as a set missing something?
    - If you had to add a 5th priority, what would it be?
    - Would you alter any of the existing priorities or their subcomponents?
  - 5 minutes of voting:
    - Group votes for top 2-3 failure/criticism combinations and 1 (or maybe 2) 5th priority proposals.
    - Idea originator for each shares with plenary.
- Plenary 35 minutes:
  - For each scenario: scenario recap by facilitator, group feedback by nominees (5 minutes each). Recorded on table on screen. Free-flowing general discussion.
- 4:00 pm Tea including Interdepartmental Surveillance Contacts (ISC)
- 4:10-5:10 pm ISC Session, plenary introduction (CSR teams collect thoughts for next round)
- 4:10 pm ISC/CSR groups (CSR groups stay as is and keep scenario; joined by ISC): 1 minute-taker, 1 process lead.
  - 10 mins: 1 CSR team member presents scenario synopsis to ISC members. Remaining 3-4 CSR team members give 1 illustration each of what goes wrong for our members because of wrong advice (ideally covering most/all surveillance priorities) from post-lunch plenary to this scenario (and drawing on table projected on the screen).
  - 5 mins: 2 CSR team members discuss/present a failure/criticism combination from previous pre-mortem.
  - 5 mins: Individual members think and write down multiple answers to these two questions: Imagine the IEO is writing a report highly critical of the CSR in 2030. (1) What criticisms will it write? (2) What are the underlying reasons for these critiques?
  - 15 mins: Go around table until all ISC (and any new CSR) ideas noted (stickies on board).
  - 15 mins: As a result, how should surveillance priorities be adjusted?
  - At 5:00 pm: Within each group, each ISC member gets 2 votes to choose how surveillance priorities should be adjusted using stickers on flipchart. Originators of 2 winning statements present in plenary.
- 5:10 pm Plenary:
  - Each table reports back on 2 changes to SPs to the other groups (10 minutes).
  - Closing comments / reactions on commonalities across scenarios (10 minutes).
- 5:30 pm Close

### Facilitation guidance (Online Annex 7: Facilitation of Scenario Planning)
- Purpose and role of facilitators:
  - External facilitators were used while developing in-house capacity (examples given).
  - In-house capacity developed via Oxford Scenarios Program, foresight conferences, and facilitation training tailored to IMF culture and work processes.
  - Early facilitator work: identify purpose and user; interrogate objectives; calibrate expectations of resources and division of labor; clarify that scenario process is facilitated, not outsourced.
- Tailoring workshops:
  - Account for human resources and deadlines; ensure drafters and good feedback providers attend; write scenarios to challenge participants.
  - Decide how much risk to take in workshops to allow unanticipated insights versus meeting stated objectives.
  - Facilitators produce scenario-planning timeline and individual workshop agendas; these must be agreed to and communicated by the project supervisor.
  - For trends workshops: define deliverables in advance and allocate time for research, brainstorming, and prioritizing.
  - For scenario planning: fit scenario building steps into available time.
  - For implications workshops: make discussion questions appropriate, striving for candor and specificity.
  - For policy gaming: identify shock(s), roles, players, and rules.
- Facilitator functions and logistics:
  - Role described as a “midwife” (Wilkinson, 2017, page 5), not the person giving birth.
  - Smooth logistics: rooms with better light and acoustics, refreshments, stationery pre-populated templates.
  - General facilitation principles:
    - (1) suggest turning off or handing in mobile phones;
    - (2) limit participants’ comment length to allow all to speak without interruption;
    - (3) emphasize that no idea is a bad idea, though any idea can be challenged, especially in brainstorming sessions;
    - (4) designate the workshops as a safe space for difficult conversations.
  - Foresight facilitation advice:
    - During trends workshops, surface assumptions by asking “Why” more than once.
    - Facilitators may act as devil’s advocates and should challenge implicit extrapolations from past/present to future.
    - During scenario-building, remind participants the process is by design uncomfortable, push to stretch plausibility, avoid scenarios being too similar, and help evaluate scenarios as a set.
    - Facilitators help project leaders make judgement calls like scenario selection and can provide assertive advice if workshops hit an impasse.
    - For implications workshops, encourage precise and frank answers; for policy-gaming, encourage precise pro and con arguments.

### Policy Game Example Instructions (Online Annex 8)
- General role-play guidance:
  - Several players represent multiple implicit roles (e.g., emerging markets with vaccine production capacity could be China, India, and Brazil).
  - Each player should adopt the character(s) of the Actor(s) through role-play, reflecting economic leverage, culture/personality, and history.
  - Enthusiasm and a thorough reading of briefing materials are sufficient preparation.
  - Players with multiple implicit countries/firms can take a single action on behalf of the collective group or pick a single implicit actor.
  - At the start of the game, players revisit a provided objective and write-down a second one. Objectives can be kept secret if preferred.
- Matrix Game mechanics: "Pros and Cons" System
  - Each argument is broken down into steps:
    - The active Player states:
      - One action (e.g., increase health spending; launch a public communications campaign; etc.).
      - The expected impact of the action.
      - Reasons (Pros) why it will be successful. Success is based on whether you will be able to carry out the action and whether it will have the intended impact.
    - Other Players then state objective reasons why it might or might NOT be successful (Additional Pros + Cons). The main player can add additional objective Cons at this time as well.
    - A required dice roll is determined using one of two methods:
      - Method 1 (used in this application): All players offer a probability of whether, based on all the pros + cons, the argument is likely to be successful. The probability (e.g., 70 percent) is then mapped to the roll required (e.g., 6 or better).
      - Method 2 (used in other games): The adjudicator groups and evaluates pro/con arguments before reducing the required roll by 1 for each pro argument and increasing the required roll by 1 for each con argument. Example: 2 pros and 1 con would translate to a required roll of 7-2+1=6.
    - Dice are rolled to determine if the required number is obtained. Luck introduces rare failures of likely successes and occasional successes of unlikely actions. Dice rolls inform narrative (marginal success, resounding success).
    - A turn ends when all players have attempted an argument.
- Negotiation and deals:
  - After the first turn or the first day, deals can be negotiated with other players (e.g., via email or otherwise).
  - At the beginning of the second session:
    - 1. The deal must be announced.
    - 2. All involved parties must verbally state their agreement and the intended specific impact of the deal for their country.
- Examples of Actions by actor type
  1. Government Actors:
    - a. Make deals with vaccine producers
    - b. Diplomatic/Military (e.g., sign agreements with other nations on vaccine purchases/joint manufacturing)
    - c. Fiscal: increased/decreased spending, raise/lower taxes, social transfer programs
    - d. Monetary: raise/lower interest rates, open market operations, CFMs, interventions, QE, macroprudential policies
    - e. Structural: governance, labor laws, business environment, price controls, PFM, SOEs, social safety nets (e.g., subsidies)
    - f. Trade: tariffs, non-tariff barriers (including on vaccines or affiliated products), currency devaluation, CFMs
    - g. Financing requests
  2. Multilaterals:
    - a. Organizing vaccine production, transport, and distribution
    - b. Surveillance
    - c. Introducing new lending facilities (e.g., pandemic loan facility)
    - d. Catalyzing other development partners’ support
    - e. Capacity development
    - f. Setting institutional views/global standards
    - g. Changing strategic focus
    - h. Convening countries, experts, etc. for discussions/negotiations
  3. Private Companies:
    - a. Allocating vaccines
    - b. Mergers and acquisitions (including hostile takeovers)
    - c. Research and Development Efforts (e.g., confirming efficacy)
    - d. Lobbying Missions to Domestic/Foreign Governments
    - e. Hiring and Recruitment Drives of new talent
    - f. Publicity Campaigns
    - g. Industrial Espionage/ Bribery
    - h. Venture Capital/IPOs or other funding rounds.
    - i. Government loans and partnerships
- Note: Actions should take advantage of country-specific factors; this applies to factors specific to certain companies or multilateral organizations.

*Source: htnea2021010-s001 - ANNEX 6.3 STRESS-TESTING DRAFT PROPOSALS USING SCENARIOS*

### 3.      In place of making an argument, each party to the deal rolls the 2 dice to determine whether or not the deal yie

### 3.      In place of making an argument, each party to the deal rolls the 2 dice to determine whether or not the deal yielded the intended benefit for that player.

### Dice-based deals and success rates
- In place of making an argument, each party to the deal rolls the 2 dice to determine whether or not the deal yielded the intended benefit for that player.
- A sum of 5 or higher is needed for success.
- Why should I make a deal? Deals benefit from a disproportionate chance of being successful (each signatory must roll a 5 or higher to receive the intended impact of the deal, with an 83% chance of success).

### Time horizon for a turn
- The length represented by a game turn may be variable during the game based on the events taking place, but typically occur over a period of about a year (i.e., giving time for trade tariffs to take effect).

### Exogenous shocks and uncertainty
- It’s a world of heightened uncertainty, so there might be some unexpected surprises.

### More on arguments — permitted scope and adjudication
- You can argue about:
  - your own abilities or that of your counterparts,
  - the existence of people, places, things or events,
  - the weather, disease or public opinion,
  - the expected effectiveness of a given action (e.g., whether conditions are ripe for a large or small fiscal multiplier).
- The actions and consequences of arguments are reflected in writing the results on the PowerPoint so the Players can keep track of what is going on.
- Clarifications on scope and common issues:
  - A player cannot physically block a policy (e.g., a new tax law) by argument, but can argue about likelihood of implementation (e.g., political will is low because of recent social unrest).
  - The only criterion for judgement is the likelihood of the event/action being successful.
  - A common error is for a player to argue that another player will be influenced or will agree to a course of action; the present player can simply be asked. Allowing time between turns enables negotiation and consent.
  - A player may argue that all parties come to negotiations; the arguing player states the case, then other players are asked if they want to come along.
  - Players are limited to one action a turn; attempts to do several different things in one turn are not allowed.
  - If two arguments are in direct opposition ("This happens" - "No it doesn't") they represent a logical inconsistency; the earlier argument has already happened, so contradicting it later is highly unlikely to succeed.
  - A player who previously failed to achieve an effect (e.g., a devaluation failed to yield export response) will be hard pressed to argue it will work the next time unless circumstances have changed; reversing prior actions tends to make for a poor narrative and should be discouraged.

### How do you win the game?
- Emphasis is placed on the process of game-playing and what comes out of the action-reaction sequence.
- Meeting one’s objectives is also a barometer of success:
  - Prior to the start of the game, players are given one and should write down another objective based on the briefing (these can be kept secret).
  - Progress is checked around halfway through the game and at the end.
- The number of times your proposed action or negotiation succeeds is another measure.

### Online Annex 9: Options for Foresight Engagement — comparison of methods (benefits, risks, resource costs)
- Full Scale, Wide Angle Scenario Planning
  - Examples: 2012-14 scenario planning round to stress-test the Fund’s business model.
  - Positives:
    - Meets periodic need to inform strategic direction.
    - Offers respite from daily activities.
    - Enhanced familiarity and ownership by user (especially top-level).
    - Experience of building scenarios is challenging and insightful, and ultimately more influential on subsequent stages.
  - Negatives/Risks:
    - Without existing demand or operational directive, value can be harder to demonstrate in advance.
    - Facilitation by junior foresight team or outside experts have drawbacks; lack of senior in-house staff may be limiting.
  - Resource Costs:
    - 3-4 days direct involvement by user possibly with top-level and senior staff though can be spread over a long period (e.g., months or years).
    - Considerable preparation from foresight team, especially for top-level events.

- Targeted Scenario Planning
  - Examples: Future of work in Africa.
  - Positives:
    - More potential for a direct output.
    - More chance of a deep dive on an issue and actionable conclusions.
    - Synergies between staff work needed for direct output and for scenario planning.
  - Negatives/Risks:
    - Can be insufficiently aware of broader factors without recent wide-angle exercise or in-built research.
    - Topic chosen, if operational, can be “loaded” and participants may hold positions rather than an open mind.
  - Resource Costs:
    - 4-8 hours direct involvement by user.
    - Low/no marginal cost on staff preparatory work on trends and uncertainties.
    - Identified project can save time on establishing objectives, but still considerable preparation from foresight team.

- Engagement with Prepared Scenarios (premortems, gamification, targeted questions)
  - Examples: Future of technical assistance on fiscal matters (based on 2017 wide-angle scenarios).
  - Positives:
    - Can leverage existing materials if the existing scenario set is still relevant.
  - Negatives/Risks:
    - Scenarios should be built to purpose; fewer topics can be addressed.
    - Less ownership and experiential learning from the scenarios.
    - Less differentiation from other exercises or publicly available scenarios.
  - Resource Costs:
    - 2-4 hours for user.
    - Can require only moderate preparation from foresight team.

- Policy-gaming (strategic simulation exercises with roleplay)
  - Examples: Capital Flow Measures.
  - Positives:
    - Experiential learning by taking on a role (including the IMF or within the IMF).
    - Helps develop contingency planning.
    - Facilitates team building.
  - Negatives/Risks:
    - Can amplify biases/priors.
    - Enjoyment of the game can cloud substantive learning.
    - Gamification of serious risks can appear dismissive of real-world stakes.
  - Resource Costs:
    - ½ day for user plus trial runs.
    - Moderately high prep needs from RU team and other staff on subject matter.
    - Careful adjudication for top-level events.

- “Unknown Unknowns” & Breaking Trends Workshop
  - Examples: Start of 2012-14 scenario planning round.
  - Positives:
    - Can be limited prep unless participants require preparatory materials.
    - Less structured format may encourage candid sharing of ideas.
    - Can be a non-committal introduction to foresight work.
  - Negatives/Risks:
    - Less likely to unearth “ah-ha” findings as creativity is endogenous to prior views.
    - Hard to structure to produce meaningful output in a short time.
    - Less use of RU expertise or techniques.
  - Resource Costs:
    - 2-4 hours with user.
    - Moderate preparation from foresight team if briefing materials can be contained.

- Pre-mortem
  - Examples: Stress-testing strategic priorities of the Comprehensive Surveillance Review (paired with scenario planning).
  - Positives:
    - Useful for highly targeted questions.
    - Resource (people & time) efficient.
    - Intuitive.
  - Negatives/Risks:
    - If participants do not find the failure scenario plausible, it can prevent useful discussions.
    - May dissuade decision-makers from taking any action.
  - Resource Costs:
    - 2 hours with users.
    - Moderate preparation from facilitators.

- Tabletop exercise (TTX) / crisis simulation
  - Examples: 2006-07 financial crisis simulation.
  - Positives:
    - Decision-makers play themselves; simulation typically occurs over multiple days with several in-game scenario updates.
  - Negatives/Risks:
    - Highly resource intensive for facilitators and participants.
    - If participants do not find the failure scenario plausible, they may not take it seriously.
  - Resource Costs:
    - 12+ hours with users.
    - Extensive preparation needs from facilitation team.

- Aspirational scenario / visioning exercises
  - Positives:
    - Can engender buy-in in contexts of inertia or risk aversion.
  - Negatives/Risks:
    - Can be seen as reflecting powerful actors’ biased agendas.
    - May appear naively optimistic or too extensive of a wish list.
  - Resource Costs:
    - 1-2 days with users depending on the adopted approach.
    - Considerable preparation from foresight team, especially for top-level events.

- Horizon scanning for tail risks
  - Examples: Tail Risk Group.
  - Positives:
    - Highlights potential emergent risks.
  - Negatives/Risks:
    - Can be viewed as a waste of constrained resources.
  - Resource Costs:
    - Resource needs vary largely.

*Source: https://www.imf.org/-/media/files/publications/analytical-notes/2021/english/htnea2021010-s001.pdf*

---


_Source: https://www.imf.org/-/media/files/publications/analytical-notes/2021/english/htnea2021010-s001.pdf_