Emerging Risks, Crisis Leadership and the Future of Risk Management

Introduction

Risk management is increasingly concerned with events that are uncertain, interconnected, rapidly evolving and capable of producing losses beyond the experience represented in historical data. Climate change, pandemics, geopolitical conflict, cyber incidents, demographic transitions and fragile supply networks can interact in ways that challenge conventional insurance and enterprise-risk models.

Traditional risk management often begins with a relatively stable exposure, estimates its probability and severity, and selects controls or financing arrangements. Emerging-risk management must go further. It must ask how the risk itself may change, how separate risks may accumulate, where feedback loops may arise, and whether existing governance arrangements can respond quickly enough.

For insurers, these developments affect underwriting, pricing, reserving, capital adequacy, reinsurance, claims operations, investments and conduct obligations. For managers, they require horizon scanning, scenario analysis, stress testing, crisis simulation and adaptive decision-making. The central proposition of this chapter is that future-ready risk management is not the prediction of a single future. It is the disciplined preparation of the organisation for several plausible futures.

Learning Objectives

By the end of this chapter, the learner should be able to:

1. Explain the characteristics of emerging, systemic and accumulating risks.
2. Evaluate climate, pandemic, geopolitical, supply-chain and demographic risks.
3. Apply horizon scanning, scenario analysis, stress testing and crisis simulation.
4. Analyse risk interdependencies, concentration and systemic accumulation.
5. Assess the role of crisis leadership, governance, communication and ethics.
6. Design adaptive and future-ready risk strategies for insurers and other organisations.

1. The Changing Risk Landscape

An emerging risk is a new or changing threat whose consequences, probability or transmission mechanisms are not sufficiently understood for conventional measurement. It may be entirely novel, such as a new technological hazard, or familiar but transformed, such as flood risk intensified by changing climate conditions.

Several characteristics distinguish emerging risks:

• Uncertainty: evidence may be incomplete and probability estimates unstable.
• Novelty: historical loss experience may be a poor guide.
• Rapidity: exposure and transmission can change faster than governance cycles.
• Interdependence: one event can activate several risk categories.
• Non-linearity: a small change in conditions can produce a disproportionately large effect.
• Ambiguity: stakeholders may disagree about causes, severity or appropriate action.
• Potential irreversibility: environmental degradation, reputational damage or institutional failure may not be quickly reversible.

A useful distinction is between risk and uncertainty. In a narrow analytical sense, risk refers to situations in which outcomes and probabilities can be estimated, while uncertainty refers to situations in which one or both are poorly known. In practice, the boundary is not fixed. A risk can move from deep uncertainty toward measurable risk as evidence accumulates, or in the opposite direction when the underlying system changes.

Risk, hazard, vulnerability and exposure

A hazard is a potentially damaging event or condition. Exposure refers to people, assets, activities or portfolios that may be affected. Vulnerability describes the susceptibility of those exposures to harm. Capacity is the ability to prevent, absorb, respond to and recover from an event.

A simple conceptual representation is:

This is not a universal actuarial identity or a precise probability equation. It is a diagnostic model. It reminds managers that risk can be reduced by lowering the hazard, reducing exposure, decreasing vulnerability or increasing response capacity. The expression is most useful for qualitative comparison; its terms are not necessarily measured in commensurable units.

Systemic and accumulation risk

Systemic risk is the possibility that stress in one part of a system will impair the functioning of the system as a whole. In insurance, systemic risk may arise when many policyholders, insurers, banks, reinsurers or financial markets are affected by the same event. Accumulation risk is the concentration of exposures that creates correlated losses, even where individual policies appear acceptable.

Examples include:

• many properties insured in the same floodplain;
• multiple insured businesses dependent on one port;
• common cyber software used by thousands of policyholders;
• life and health claims arising from one epidemic;
• insurers holding similar assets and selling similar products;
• reinsurance programmes relying on the same limited global capacity.

Diversification reduces risk only when exposures are not highly correlated. A portfolio that appears geographically diversified may still be concentrated by supplier, technology, climate peril or economic sector.

2. Major Emerging Risks

2.1 Climate change and transition risk

Climate change creates both physical and transition risks. Physical risks arise from changing weather patterns and climate-related hazards, including floods, heatwaves, storms, drought, wildfire, sea-level rise and water stress. These may be acute, such as a cyclone, or chronic, such as a gradual increase in average temperature.

Transition risks arise from movement toward a lower-carbon economy. They include changes in regulation, carbon prices, technology, consumer preferences, litigation, energy markets and asset values. An energy-intensive company may face declining demand, stranded assets or higher financing costs even if it has not suffered direct physical damage.

Climate risk is challenging because it is:

• location-specific but economically interconnected;
• cumulative rather than limited to one event;
• affected by adaptation and public infrastructure;
• subject to model uncertainty and non-stationary historical data;
• relevant to both liabilities and investments.

Insurance implications include changing catastrophe models, underwriting restrictions, premium adequacy, exclusions, parametric products, resilience incentives and the treatment of climate-related litigation. Insurers must also consider whether their investment portfolios are exposed to carbon-intensive assets or regions vulnerable to physical damage.

Climate scenario analysis should not be confused with a forecast. A scenario is a coherent representation of possible future conditions. It may examine different temperature pathways, adaptation assumptions, regulatory speeds and economic responses. The objective is to understand vulnerability and strategic options, not to claim certainty about one outcome.

2.2 Pandemics and biological threats

Pandemics combine health, operational, economic, legal and social risks. They can generate simultaneous claims, employee absence, supply interruptions, travel restrictions, business interruption and financial-market stress. Unlike a local catastrophe, a pandemic may affect policyholders, insurers, reinsurers, service providers and governments at the same time.

Key analytical issues include:

• transmission speed and geographic spread;
• uncertainty concerning severity and variants;
• changing public-health interventions;
• claims triggers and policy wording;
• accumulation across life, health, travel, event cancellation and business interruption lines;
• dependence on outsourced claims and customer-service operations.

Pandemic preparedness requires more than a business-continuity plan. It requires alternative operating locations, remote-work capability, critical-staff substitution, data security, medical and public-health liaison, liquidity planning and clear escalation thresholds.

2.3 Geopolitical disruption

Geopolitical risk includes interstate conflict, sanctions, trade restrictions, political violence, diplomatic breakdown, cyber operations, election-related instability and abrupt regulatory change. Its effects may be direct, such as war damage, or indirect, such as energy-price shocks, capital controls or restricted access to technology.

For insurers, geopolitical disruption affects political-risk insurance, marine and aviation insurance, credit risk, investment portfolios, reinsurance recoverability and the legality of transactions. Sanctions compliance is especially important: a claim payment, investment or reinsurance arrangement may become prohibited after a political event.

Geopolitical risk is often characterised by threshold effects. A long period of tension may produce little measurable loss until a triggering event causes rapid escalation. Managers should therefore monitor indicators such as military mobilisation, diplomatic signals, sanctions proposals, shipping patterns, commodity prices and cyber activity, while recognising that indicators are not reliable predictions.

2.4 Supply-chain fragility

Supply chains are vulnerable when they are highly concentrated, opaque, geographically exposed or dependent on critical nodes. A single supplier may be replaceable in theory but irreplaceable within the time required by the business. Fragility can arise from just-in-time inventories, limited transport capacity, scarce components, port dependence, cyber vulnerabilities and weak visibility beyond tier-one suppliers.

A supply chain can be represented as a network rather than a linear sequence. Failure at a central node may affect many firms simultaneously. Important risk-management measures include supplier mapping, criticality classification, dual sourcing, strategic inventory, alternative logistics routes, contractual rights to information, supplier financial monitoring and cyber assurance.

Insurance products may cover contingent business interruption, trade credit, marine cargo, political risk and supply-chain disruption. However, coverage must be aligned with actual triggers, waiting periods, sublimits, exclusions and proof-of-loss requirements. Risk transfer cannot substitute for understanding the network.

2.5 Demographic shifts

Ageing populations, migration, urbanisation, changing household structures and workforce transformation influence risk over long time horizons. Demographic change affects longevity, health expenditure, pension liabilities, labour availability, property demand, caregiving needs and product design.

For life insurers and pension providers, longevity risk arises when policyholders live longer than expected. Mortality improvements may increase the present value of annuity obligations. Conversely, unequal health outcomes and changes in employment may increase morbidity or lapse risk. Demographic assumptions must therefore be regularly reviewed rather than treated as permanent parameters.

Demographic change also creates distributional and ethical questions. Products may become unaffordable for vulnerable groups, while automated underwriting may reproduce historical inequalities. Fairness, accessibility and transparency are essential components of responsible innovation.

3. Tools for Anticipating and Testing Emerging Risk

3.1 Horizon scanning

Horizon scanning is the systematic exploration of emerging trends, weak signals, discontinuities and plausible threats. It should be connected to decisions rather than conducted as an isolated research exercise.

A practical process is:

1. Define the scope, time horizon and decision context.
2. Collect signals from internal data, regulators, academia, industry, technology specialists and affected communities.
3. Classify signals by severity, uncertainty, speed and interdependence.
4. Challenge assumptions through multidisciplinary workshops.
5. Develop scenarios and identify leading indicators.
6. Assign ownership and establish review dates.
7. Link findings to capital, underwriting, continuity and strategic decisions.

Weak signals should not automatically become forecasts. Their value lies in stimulating inquiry and reducing surprise. A governance board may use a risk radar that displays risks by impact and preparedness, but visual placement should not create false precision.

3.2 Scenario analysis

Scenario analysis examines how an organisation might perform under internally coherent descriptions of future conditions. Scenarios may be exploratory, testing what could happen, or normative, describing a desired future and pathways toward it.

A robust scenario contains:

• a time horizon;
• key external drivers;
• assumptions about policy, technology and behaviour;
• causal relationships;
• operational and financial consequences;
• management responses;
• indicators that would suggest the scenario is becoming more plausible.

Scenarios should include both severe plausible events and slow-burn developments. A scenario that is merely extreme but implausible may distract from more material vulnerabilities.

3.3 Stress testing

Stress testing evaluates the effect of specified adverse conditions on solvency, liquidity, operations or strategic objectives. It differs from ordinary forecasting because it deliberately examines adverse but plausible conditions.

A simple loss aggregation model illustrates the importance of dependence. If total loss is the sum of component losses:

where L is total loss and L_i is loss from exposure or risk component i. The expected total is:

However, variability depends on covariance:

Here Var denotes variance and Cov(L_i,L_j) denotes covariance between components. If losses are positively correlated, the covariance terms increase aggregate volatility. The formula assumes finite variances and a well-defined joint distribution; it does not require normality, but estimation of the covariance terms requires credible data or expert assumptions.

For two losses, with standard deviations σ₁ and σ₂ and correlation ρ,

If ρ = 0, the losses are uncorrelated, not necessarily independent. If ρ > 0, aggregation produces greater volatility than the independent case. In a crisis, correlations may rise, making historical diversification assumptions unreliable.

Stress tests should examine solvency, liquidity and operational capacity separately. An organisation may remain solvent on an accounting basis but fail because it cannot meet immediate cash demands, access essential staff or process claims.

3.4 Reverse stress testing

Reverse stress testing begins with a failure outcome and asks what combination of conditions could produce it. Examples include breach of a capital threshold, inability to pay claims, loss of a critical licence or prolonged service interruption.

The method is valuable because it exposes hidden assumptions. Management may discover that failure does not require one catastrophic event; it may result from several moderate shocks occurring together, such as rising claims, collateral calls, asset-market losses and operational absenteeism.

3.5 Crisis simulation and exercising

A crisis simulation tests decisions, communications, information flows and authority under time pressure. A well-designed exercise includes realistic injects, incomplete information, conflicting priorities and changing conditions.

Exercises should assess:

• whether escalation criteria are understood;
• whether decision rights are clear;
• whether data and dashboards are timely and reliable;
• whether critical suppliers can be contacted;
• whether customer and regulator communications are consistent;
• whether liquidity and claims processes function;
• whether lessons are recorded and implemented.

A simulation is not successful merely because participants complete the exercise. Its value depends on corrective action, ownership and retesting.

4. Crisis Leadership and Governance

Crisis leadership differs from routine management. Leaders must make consequential decisions with incomplete information, preserve legitimacy, coordinate specialists and maintain organisational purpose. They must avoid both paralysis and unjustified certainty.

Effective crisis leaders:

• establish priorities based on life safety, legal duties, customer protection and organisational viability;
• distinguish facts, assumptions and unknowns;
• create a manageable decision rhythm;
• delegate authority while retaining accountability;
• communicate what is known, unknown and being done;
• protect challenge and dissent;
• revise decisions as evidence changes;
• maintain a record of decisions and reasons.

The board should oversee risk appetite, resilience investment, succession, crisis protocols and material exposures. Senior management should operationalise these requirements. The chief risk officer should have access to the board and sufficient independence to challenge commercial pressure.

Risk culture is revealed during stress. A culture that rewards bad-news reporting, scenario challenge and transparent escalation is more resilient than one that treats uncertainty as managerial failure.

Communication and stakeholder trust

Crisis communication is an operational control, not merely a public-relations activity. Inconsistent or delayed communication can increase litigation, reputational loss and customer harm. Messages should be accurate, accessible, proportionate and coordinated with regulators and public authorities.

Stakeholders include policyholders, employees, suppliers, regulators, investors, communities and the media. Ethical communication avoids speculation, concealing material facts or shifting blame to affected parties. Organisations should explain uncertainty without using it as an excuse for inaction.

5. Adaptive Strategy for Insurers

A future-ready insurer combines prevention, resilience, risk transfer and financial capacity. Key practices include:

• updating underwriting guidelines as exposures change;
• monitoring accumulation by peril, geography, supplier and technology;
• using multiple models and expert judgment rather than relying on one model;
• incorporating forward-looking climate and catastrophe information;
• testing reinsurance protection under correlated events;
• maintaining liquidity buffers and contingent funding;
• investing in claims automation while preserving human review;
• developing parametric and resilience-linked products where triggers are transparent;
• reviewing exclusions and wordings for clarity and legal robustness;
• integrating sustainability, conduct and prudential considerations;
• establishing data lineage, model validation and escalation standards.

Technology can improve horizon scanning and early warning. Artificial intelligence may identify patterns in text, satellite data, claims and supplier information. However, AI outputs may reflect biased data, obscure causal reasoning or create false confidence. Human oversight, explainability, cybersecurity, privacy protection and model-risk governance remain essential.

A useful strategic cycle is:

Figure: An adaptive risk-management cycle linking intelligence, testing, action and learning.

6. Illustrative Case: Coastal Insurer Facing Compound Risk

Background

Harbour Mutual is a fictional regional property insurer operating in a coastal economy. Its portfolio includes residential property, small businesses and marine-related enterprises. Historical underwriting has relied on long-term catastrophe averages and relatively stable reinsurance pricing.

Problem

Over three years, the region experiences severe rainfall, a port closure caused by a cyber incident and an unusually intense cyclone. The events are not independent: flooding damages roads, the port closure delays building materials, and reconstruction costs increase claims severity. Reinsurance capacity tightens, while investment losses reduce available capital.

Analysis

The insurer’s individual property risks appeared acceptable, but accumulation analysis revealed concentration in two coastal districts and dependence on one major port. Claims data also showed that repair inflation and contractor scarcity were increasing settlement costs. A conventional catastrophe model understated the combined effect because it treated supply disruption and physical damage separately.

A reverse stress test identified a failure pathway involving: a major cyclone, delayed reinsurance recoveries, a 20% increase in repair costs, and a short-term liquidity outflow. The board concluded that solvency alone was not the immediate constraint; claims-processing capacity and liquidity were more urgent.

Approach and solution

Harbour Mutual implemented the following measures:

1. It mapped exposure by flood depth, infrastructure dependence and supplier concentration.
2. It introduced accumulation limits for highly correlated coastal zones.
3. It negotiated alternative claims suppliers and emergency procurement arrangements.
4. It established a liquidity trigger linked to projected claims payments and collateral calls.
5. It conducted a simulation involving the insurer, reinsurer, regulator, port authority and major vendors.
6. It offered resilience discounts for approved flood protection and elevated equipment.
7. It revised scenarios to include compound physical, cyber and supply-chain events.
8. It created a crisis communication protocol for policyholders and public authorities.

Outcome

The measures did not eliminate catastrophe losses. They reduced response time, improved claims communication, diversified operational dependencies and gave the board earlier warning of liquidity pressure. The insurer also identified some properties for which risk transfer was no longer economically sustainable without mitigation or public-private support.

Learning points

The case demonstrates that emerging risk is often a problem of interaction rather than novelty. Historical data remained useful for some components, but it was insufficient for the combined event. It also shows that resilience decisions may require underwriting, operations, finance, technology, public policy and customer engagement to act together.

7. Limitations, Ethics and Regulation

Emerging-risk analysis has unavoidable limitations. Models may be misspecified, data may be biased, correlations may change and scenarios may omit unknown unknowns. Excessive reliance on quantitative outputs can create model complacency. Conversely, purely qualitative judgment may become inconsistent or vulnerable to dominant personalities.

Good practice combines models, expert judgment, historical evidence, external research and challenge. Material assumptions should be documented, sensitivity-tested and approved at an appropriate level.

Ethical issues include:

• fairness in pricing and access to insurance;
• use of personal, health and location data;
• responsibility for communities exposed to climate hazards;
• transparency of automated decisions;
• treatment of vulnerable customers during crises;
• conflicts between short-term profitability and long-term resilience;
• distribution of losses between private insurers, governments and citizens.

Regulators increasingly expect forward-looking governance, operational resilience, climate-risk assessment, adequate capital and clear customer treatment. Requirements vary by jurisdiction, but the direction is consistent: firms should be able to identify material exposures, withstand disruption, recover critical services and demonstrate responsible governance.

Best-Practice Checklist

A mature emerging-risk programme should:

• maintain a multidisciplinary risk inventory;
• distinguish hazards, exposures, vulnerabilities and capacities;
• monitor weak signals and leading indicators;
• assess dependencies and common concentrations;
• use scenarios, stress tests and reverse stress tests;
• test liquidity and operational continuity as well as solvency;
• assign clear owners and escalation thresholds;
• exercise crisis plans under realistic conditions;
• communicate honestly and inclusively;
• validate models and challenge assumptions;
• convert lessons into funded actions;
• review strategies as conditions change.

Chapter Summary

Emerging risks challenge conventional risk management because they are uncertain, interconnected, dynamic and often non-linear. Climate change creates physical and transition risks; pandemics produce simultaneous health, operational and financial stress; geopolitical events can rapidly alter markets and legal conditions; fragile supply chains transmit disruption through critical nodes; and demographic change reshapes insurance liabilities and social needs.

Systemic and accumulation risk require attention to correlation, common dependencies and portfolio concentration. Horizon scanning identifies weak signals and changing drivers. Scenario analysis explores plausible futures, while stress testing evaluates adverse conditions and reverse stress testing works backward from failure. Crisis simulations test whether plans, authority, information and communications function under pressure.

Future-ready risk leadership combines technical analysis with judgment, ethics, governance and adaptive learning. Insurers must connect underwriting, capital, reinsurance, operations, technology, investments and stakeholder communication. No model can remove uncertainty, but disciplined preparation can reduce surprise, improve resilience and support responsible decisions.

Key Terms

Emerging risk: A new or changing risk whose consequences or probability are insufficiently understood.

Systemic risk: Risk that disruption in one part of a system impairs the system as a whole.

Accumulation risk: Concentration of correlated exposures that can produce large aggregate losses.

Hazard: A potentially damaging event or condition.

Exposure: A person, asset, activity or portfolio subject to a hazard.

Vulnerability: Susceptibility of an exposure to harm.

Response capacity: Ability to prevent, absorb, respond to and recover from disruption.

Physical climate risk: Loss arising from climate-related physical events or gradual changes.

Transition risk: Loss arising from movement toward a different economic, regulatory or technological system.

Horizon scanning: Systematic exploration of trends, weak signals and possible disruptions.

Scenario analysis: Examination of consequences under coherent descriptions of possible futures.

Stress testing: Assessment of performance under specified adverse conditions.

Reverse stress testing: Analysis that begins with a failure outcome and identifies causes.

Crisis simulation: An exercise testing decisions, coordination and communications under pressure.

Non-linearity: A relationship in which changes in inputs do not produce proportional changes in outputs.

Correlation: A measure of the direction and strength of linear association between variables.

Operational resilience: Capacity to continue and recover important services during disruption.

Risk appetite: The amount and type of risk an organisation is willing to accept.

Weak signal: An early, uncertain indication of a potentially important development.

Adaptive strategy: A strategy designed to change as evidence, conditions and feedback evolve.

Review Questions

Short Answer Questions

1. What distinguishes an emerging risk from a conventional, well-understood risk?
2. Explain the difference between systemic risk and accumulation risk.
3. Identify two physical and two transition risks associated with climate change.
4. What is the purpose of reverse stress testing?
5. Why is crisis communication considered an operational control?

Long Answer Questions

1. Discuss how pandemics can create simultaneous insurance, operational and financial risks.
2. Explain the role of horizon scanning in a future-ready risk-management system.
3. Analyse why historical catastrophe data may be insufficient for climate-related decision-making.
4. Discuss the governance and ethical responsibilities of insurers using artificial intelligence in risk assessment.
5. Explain how supply-chain fragility can create accumulation risk for an insurer.

Analytical / Application Questions

1. Design a stress-testing programme for an insurer exposed to a compound event involving flood, cyber disruption and supplier failure. Specify assumptions, outputs and governance responsibilities.
2. A life insurer experiences higher-than-expected longevity, falling interest rates and a shortage of skilled claims staff. Analyse the interactions among these risks and recommend adaptive responses.
3. Using the Harbour Mutual case, evaluate whether resilience incentives, accumulation limits or additional reinsurance would provide the strongest strategic benefit. Justify your conclusion and identify trade-offs.

Multiple Choice Questions with Answers

Question 1

Which characteristic most clearly distinguishes an emerging risk?

A. Its probability and consequences are fully stable and known

B. Its effects or transmission mechanisms are changing and incompletely understood

C. It can only arise from a natural catastrophe

D. It is always excluded from insurance contracts

Correct Answer: B. Its effects or transmission mechanisms are changing and incompletely understood

Question 2

Which statement best describes transition risk?

A. Damage caused directly by a cyclone

B. Loss arising from movement toward a different regulatory, technological or economic system

C. The probability that a policyholder will lapse

D. The cost of processing a routine claim

Correct Answer: B. Loss arising from movement toward a different regulatory, technological or economic system

Question 3

If two loss components have positive covariance, what is the usual effect on aggregate loss variance?

A. It reduces aggregate variance below both component variances

B. It has no effect on aggregate variance

C. It increases aggregate variance relative to the zero-covariance case

D. It proves that the components are independent

Correct Answer: C. It increases aggregate variance relative to the zero-covariance case

Question 4

What is the principal purpose of horizon scanning?

A. To guarantee an accurate prediction of the next crisis

B. To identify trends, weak signals and possible discontinuities for decision-making

C. To replace all actuarial models

D. To determine the exact annual premium for every policy

Correct Answer: B. To identify trends, weak signals and possible discontinuities for decision-making

Question 5

Reverse stress testing begins by asking:

A. What is the average historical loss?

B. Which product has the highest sales?

C. What conditions could lead to a defined failure outcome?

D. How can all uncertainty be removed?

Correct Answer: C. What conditions could lead to a defined failure outcome?

Question 6

Why can a geographically diversified portfolio still have accumulation risk?

A. Geography never affects insurance risk

B. All policies have identical claims

C. Exposures may share a common supplier, technology, peril or infrastructure dependency

D. Diversification always increases correlation

Correct Answer: C. Exposures may share a common supplier, technology, peril or infrastructure dependency

Question 7

Which is an example of operational resilience?

A. Assuming suppliers will never fail

B. Maintaining the ability to continue critical services during disruption

C. Eliminating every possible risk

D. Using only historical averages for planning

Correct Answer: B. Maintaining the ability to continue critical services during disruption

Question 8

A crisis leader should generally:

A. Communicate only when all facts are certain

B. Hide uncertainty to preserve confidence

C. Separate known facts, assumptions and unknowns while maintaining a decision rhythm

D. Delegate all accountability to technical specialists

Correct Answer: C. Separate known facts, assumptions and unknowns while maintaining a decision rhythm

Question 9

Which issue is especially important when using AI in underwriting?

A. The assumption that automated outputs are always neutral

B. Model bias, explainability, privacy and human oversight

C. Avoiding all data validation

D. Removing regulatory review

Correct Answer: B. Model bias, explainability, privacy and human oversight

Question 10

What did the Harbour Mutual case primarily demonstrate?

A. Compound events can create losses through interactions among physical, cyber and supply-chain risks

B. Reinsurance completely eliminates accumulation risk

C. Historical averages always capture future compound events

D. Claims operations are unrelated to liquidity risk

Correct Answer: A. Compound events can create losses through interactions among physical, cyber and supply-chain risks