Pandemic and Emergency Preparedness Economics

Why this matters in health economics

Most of this book is about allocating a bounded budget across recurring, well-measured health problems: a screening programme, a new medicine, a service redesign. Pandemic prevention and emergency preparedness break the frame. The events in question are rare, so the evidence base is thin; catastrophic, so the losses are society-wide rather than confined to a patient group; and correlated, so they strike every part of the system at once rather than averaging out across a portfolio. A method calibrated on the ordinary will misprice the extraordinary, and it will misprice it in a predictable direction — downward.

The stakes are not abstract. A severe pandemic is simultaneously a mortality event, a fiscal event, and an economic event: excess deaths, an overwhelmed health system, a contracting economy, and emergency public borrowing that dwarfs the cost of the preparedness that was declined. The recent global experience of COVID-19 (from 2020) made the asymmetry vivid — the sums spent responding to the emergency exceeded, by orders of magnitude, the sums that had been declined for preparedness beforehand, though the precise multiple depends on which costs are counted and over what horizon. The direction of the lesson is not in doubt even where the exact figures are contested.

For a director, three practical stakes follow. First, timing: preparedness must be bought before the event, when the threat is invisible and the queue of visible, fundable health needs is long. Second, appraisal: standard cost-effectiveness analysis, built on expected values and stable probabilities, struggles to value an option that pays off only in a rare state of the world. Third, politics: attention and money surge during a crisis and evaporate after it — the "panic-and-neglect" cycle — so any preparedness commitment must be designed to survive the neglect phase, when it is cheapest to cut and hardest to defend.

Core concepts

Low-probability, high-consequence events and tail risk. Preparedness concerns the tail of the loss distribution: outcomes that are individually unlikely in any given year but carry enormous consequences if they occur. Tail risk is the risk of these extreme, low-frequency events, and it is exactly where ordinary intuition and ordinary appraisal fail. When losses are "fat-tailed" — the extremes are far larger and less improbable than a normal distribution would suggest — the average outcome is a poor guide, because a single realization in the tail can exceed the sum of all the years in which nothing happened.

Expected value and why it under-counts catastrophe. The standard decision rule multiplies each outcome by its probability and sums: an expected value calculation. It is the workhorse of economic evaluation, but it has two weaknesses for catastrophe. First, the probability of a novel pandemic in any given year is small and poorly estimated, so the expected annual loss looks modest and preparedness looks expensive per unit of expected benefit. Second, expected value is risk-neutral: it treats a certain loss and a gamble with the same mean as equivalent, whereas societies are risk-averse to ruin. Where the loss threatens the system itself, expected-utility reasoning — which weights bad states more heavily — justifies paying more than the actuarially "fair" premium to avoid the tail.

Option value and preparedness as a real option. Much of preparedness buys not a service but the ability to act if an emergency arrives: standing manufacturing capacity, a stockpile, a trained surge workforce, a research pipeline. This is a real option — an asset whose worth lies in the flexibility to respond, which conventional net-present-value arithmetic, comparing expected costs to expected benefits, tends to miss. The related option value in cost–benefit analysis captures what a decision-maker would pay to keep a future course of action open under uncertainty. Preparedness is, economically, a portfolio of such options: each stockpile or platform is a call option on a rapid response, valuable precisely because the future is uncertain.

Surge capacity. Surge capacity is the ability of a health system to expand rapidly beyond its normal capacity to meet a sudden rise in demand — beds, staff, ventilators, laboratory throughput, and the space and logistics to use them. It is deliberately slack: capacity that sits idle in normal times and earns its keep only in an emergency. That idleness is why annual efficiency drives erode it, and why its value has to be argued on option-value rather than utilization grounds.

The value of preventing catastrophe. Valuing preparedness requires putting a figure on mortality and on economic disruption averted. The value of a statistical life — the marginal rate of substitution between money and fatality risk, used across transport, environmental, and health appraisal — lets analysts monetize avoided deaths. But a pandemic's losses are not only deaths: they include the output lost to lockdowns, the learning lost to closed schools, and the second-order harms of a health system consumed by one disease. Standard appraisal handles independent, marginal risks well and correlated, society-wide, non-marginal ones badly.

Public goods and the under-provision problem. Preparedness has strong public-good features — surveillance, pathogen genomics, and response capacity are non-rival and hard to exclude — so both markets and national budgets under-provide them; the cross-border dimension of this is developed in Chapter 4.2 — Global Health and Trade. Domestically, the failure is compounded by the annual budget cycle: preparedness competes each year against visible needs and loses, because its benefit is a reduced probability of a future event that may not arrive on any given minister's watch.

The precautionary principle and catastrophic thresholds. Where a threat is potentially irreversible and catastrophic, the precautionary principle argues for protective action even under deep uncertainty about the probability. It is not a licence to spend without limit; rather, it shifts the burden of proof when the downside is ruin. Preparedness sits alongside other global catastrophic risks — events that could inflict damage on a planetary scale — for which the same appraisal difficulties apply: the probabilities are contested, the consequences are vast, and waiting for definitive evidence may mean waiting until it is too late.

Best practices

1. Appraise preparedness as insurance and option value, not as expected cost per QALY. The workhorse ratio of cost per quality-adjusted life year (QALY), used across Chapter 2.1 — Economic Evaluation, assumes stable probabilities and marginal, independent effects — none of which hold for catastrophe. Frame the question as: what would a risk-averse society pay in premium to reduce the probability and severity of a ruinous, correlated loss? Report the option value of keeping a response capability open, not only the discounted expected benefit, and be explicit that a favourable expected-value ratio is not the bar preparedness must clear.

2. Model the tail, not the mean. Build the loss distribution and interrogate its extremes rather than collapsing everything to an expected annual figure that will always look small. Use scenario analysis across plausible severities — a moderate, a severe, and a catastrophic pandemic — and show decision-makers the tail losses directly, because a single tail realization can exceed decades of the years in which nothing happens. Where the distribution is fat-tailed, say so, and warn that historical averages understate both the probability and the size of the extreme.

3. Buy a portfolio of standing capabilities, and value the flexibility explicitly. Stockpiles, surge workforce, laboratory networks, standing vaccine and manufacturing platforms, and a live research and development pipeline are each real options on a rapid response. The 100 Days Mission — the international goal of compressing the time from identifying a novel pathogen to deploying safe, effective countermeasures — is a concrete example of paying now to shorten response later. Value the flexibility, not just the expected use, and diversify: correlated threats are best met by a spread of options rather than a bet on one scenario.

4. Design commitments to survive the neglect phase. The panic-and-neglect cycle predictably strips preparedness budgets once a crisis fades, so protect the commitment structurally rather than relying on memory. Multi-year ring-fenced funding, a standing statutory body, replenishment schedules written into law, and pre-negotiated standby contracts all convert a fragile annual line into a durable capability. Assume that political attention will lapse, and build the institution that keeps buying the premium when no one is frightened.

5. Keep stockpiles as a managed, rotating portfolio — not a warehouse that expires. A stockpile is only an option if its contents are usable when called. Manage shelf life, rotate stock into routine use where possible, hold a mix of finished goods and the capacity to manufacture, and cost the ongoing carrying and replacement burden honestly rather than treating the stockpile as a one-off purchase. A warehouse of expired or wrong-specification supplies is a sunk cost masquerading as preparedness.

6. Value surge capacity on option grounds, and defend it against efficiency drives. Slack — empty beds, cross-trained staff who can be redeployed, idle laboratory throughput — is by construction under-utilized in normal times, so ordinary efficiency metrics mark it for cutting. Make the option-value case explicitly: the capacity earns its return in the rare state of the world, and eliminating it raises the loss when that state arrives. Where full standing slack is unaffordable, buy the option more cheaply through convertible facilities, reserve contracts, and trained-but-not-deployed workforce rosters.

7. Invest at the source and upstream, where the return is highest. Containing an outbreak before it spreads, and preventing spillover at the human–animal interface, is usually far cheaper than responding to a full-blown pandemic — the return on early detection and rapid containment dwarfs the return on late response. Fund surveillance, genomic sequencing, and rapid-response teams as the front line, and treat domestic border defence as a last resort rather than a strategy. The economic logic mirrors prevention elsewhere in health: the upstream pound buys more than the downstream one.

8. Use market-shaping to pull forward capability the market will not supply alone. Private firms will not hold idle vaccine capacity or invest in countermeasures for pathogens that may never emerge, because the private return is negative in expectation. Public instruments — advance market commitments, standing agreements with the Coalition for Epidemic Preparedness Innovations (CEPI), volume guarantees, and warm-base manufacturing subsidies — pay to keep capacity ready; the innovation-incentive mechanics are developed in Chapter 5.1 — Innovation Health Economics. Structure these so the public pays for readiness and speed, not only for doses delivered after the fact.

9. Invoke the precautionary principle with discipline, not as a blank cheque. For irreversible, catastrophic threats, act under uncertainty rather than waiting for probabilities you will never have in time — but bound the spend. Set a proportionate preparedness budget sized to the plausible tail loss and the cost of capability, prioritize no-regret measures that pay off across many scenarios (surveillance, flexible manufacturing, trained workforce), and revisit as evidence accrues. Precaution justifies acting early; it does not excuse acting without a portfolio logic.

10. Count the whole social loss, and name what standard appraisal omits. A pandemic's cost is not only clinical: it is lost output, closed schools, mental-health harm, deferred routine care, and the deaths and disability those cause. Use the value of a statistical life to monetize avoided fatalities, but state plainly that correlated, society-wide, non-marginal losses sit beyond the comfortable range of cost-effectiveness analysis, and present them as scenarios rather than forcing a single point estimate. An honest appraisal is transparent about the losses it cannot fully quantify.

11. Rehearse, measure decay, and treat the response system as a perishable asset. A plan that is never exercised is a document, not a capability. Run simulations and live exercises, measure how fast trained staff, supplier relationships, and institutional knowledge decay between events, and fund the maintenance that keeps the option alive. The International Health Regulations require states to build and maintain core capacities for detection and response; treat those obligations as a floor to be exercised, not a form to be filed.

12. Pre-commit the hard allocation rules before the crisis, not during it. Scarce ventilators, intensive-care beds, and first vaccine doses will have to be rationed in a severe event, and deciding the rule under pressure invites both unfairness and paralysis. Agree the priority-setting framework in advance, transparently and with the procedural fairness that Chapter 3.3 — Rationing develops, so that the emergency executes a considered rule rather than improvising one. Preparedness includes the ethics and the logistics of saying no at scale.

Questions to discuss with your team

1. How much should we pay in premium for a catastrophe that may never happen on our watch — and how do we defend that number when the money could fund visible care today? This is the central tension of preparedness: the benefit is a reduced probability of a rare future event, while the opportunity cost is concrete, present, and easy to name. An expected-value calculation will almost always make preparedness look expensive per unit of expected benefit, because the probability in any single year is small; leaning only on that calculation guarantees under-provision. The honest angles are to frame the spend as an insurance premium against a fat-tailed loss, to size it against the plausible tail rather than the mean, and to identify the no-regret measures that pay off across many scenarios so the premium is not wasted even if the catastrophe never arrives. You should also be candid about risk aversion: a society rationally pays more than the actuarially fair premium to avoid ruin, and that is a value judgement to make explicitly, not to smuggle in. A good answer states the premium, the tail loss it insures against, and why a risk-averse public would accept the trade — not a false claim that preparedness is "cost-effective" on the ordinary ratio.

2. What will actually stay funded once the fear fades — and how do we build for the neglect phase now? Attention and money surge during an emergency and collapse after it, so the real design problem is not what to buy at the peak but what survives the trough. The tension is that the neglect phase is exactly when preparedness is cheapest to cut and hardest to defend, and when the people who lived through the last crisis have moved on. Concrete angles: which commitments are protected by statute, multi-year ring-fencing, or standing contracts rather than by an annual budget line that competes afresh each year? What is the replenishment and maintenance schedule, and is it automatic or discretionary? Who owns the capability when public attention is elsewhere? An honest answer assumes the political will to lapse and points to the institutional structures — a standing agency, legislated funding, pre-negotiated standby capacity — that keep the premium paid when no one is afraid, rather than trusting that the lesson will be remembered.

3. When we have to ration ventilators, beds, or first vaccine doses in a severe event, whose rule are we executing — and did we agree it before the crisis or during it? A severe emergency forces society-wide rationing under time pressure, and the choice is whether that rationing follows a considered, transparent rule or an improvised one made in exhausted haste. The tension is between the desire to keep options open and the reality that deferring the decision means making it worse, later, and less fairly. Concrete angles: is there a pre-agreed allocation framework, is it procedurally fair and publicly defensible, and has it been tested against realistic scenarios? Who is accountable for it, and how does it handle equity so that scarce resources do not simply track existing advantage? What is the trigger that switches the system from normal to crisis standards of care? A good answer treats allocation ethics as part of preparedness — decided, published, and rehearsed in advance — rather than as something to be worked out when the intensive-care unit is already full.

4. Are we spending at the source, where an outbreak is cheapest to stop, or at the border, where it is most visible but least effective? The economic logic of preparedness points upstream: surveillance, genomic sequencing, and rapid containment at the point of spillover avert the tail rather than merely coping with it, and the return on early detection dwarfs the return on late, downstream response. The tension is that source-level investment is unglamorous, often overseas or at the human–animal interface, and generates benefits that are hard to attribute to any one budget, while border defence is politically legible and feels like control. Concrete angles: what share of the preparedness budget funds detection and containment versus response and border measures, and can we defend that split on return grounds? Do we treat surveillance networks and rapid-response teams as the front line, or as the first thing cut when money is tight? How do we account for the cross-border benefits our surveillance generates, which our own budget will never fully capture? An honest answer ranks the cheap, upstream, no-regret measures ahead of the expensive downstream ones, and is candid that the most cost-effective components are usually the least visible.

5. How much should we pay private firms to hold capacity we hope never to use — and what exactly are we buying: readiness, speed, or doses? Markets will not hold idle vaccine lines or invest in countermeasures for pathogens that may never emerge, because the private return is negative in expectation, so public money must pull that capacity forward through advance market commitments, warm-base subsidies, and standing agreements. The tension is that paying for readiness means paying for something that produces nothing in normal times, and it is easy to structure the contract so the public bears the risk while the firm captures the upside. Concrete angles: are we paying for readiness and speed — a shortened time from pathogen to countermeasure — or only for doses delivered after an outbreak, when the option has already lapsed? How do we price standby and reserve capacity so it is a fair premium rather than a subsidy, and who owns the capacity between events? What stops the arrangement from decaying in the neglect phase, when the warm base looks like waste? A good answer names precisely what readiness is being bought, ties payment to speed and availability rather than only to output, and structures the incentive so the public pays for the option, not for a rescue after the fact.

6. A pandemic's cost is far more than clinical — so what are we leaving out of the number, and how do we present what we cannot honestly reduce to a single figure? The losses from a severe pandemic include lost output, closed schools and the learning forgone, deferred routine care, mental-health harm, and the deaths and disability those second-order effects cause — losses that are correlated, society-wide, and non-marginal, exactly where cost-effectiveness analysis is least at home. The tension is the pull to produce one tidy cost-per-QALY figure for a budget committee, when the largest and most important losses sit beyond the range that any single ratio can honestly express. Concrete angles: which losses are we monetizing — avoided deaths via the value of a statistical life — and which are we naming but not collapsing into the ratio? Are we presenting the society-wide losses as severity scenarios rather than forcing a false point estimate? How do we stay transparent about the harms we cannot fully quantify without letting that uncertainty become an excuse to value them at zero? An honest answer counts the whole social loss, states plainly what standard appraisal omits, and presents the uncomfortable, unquantifiable tail as scenarios rather than hiding it inside a spuriously precise number.

In practice: a health economics example

The Republic of Lentara is a fictional upper-middle-income country of 22 million people with a mixed tax-and-social-insurance health system. Five years after the last global pandemic, its Treasury and Ministry of Health convene a joint appraisal of a proposed National Preparedness Portfolio. The memory of the last emergency is fading, competing health needs are loud, and the finance ministry's instinct is that the country "already has a plan on file". The health minister asks the newly-established preparedness unit for an appraisal that will hold up in a budget committee that thinks in cost per QALY.

The portfolio has four components: a rotating stockpile of personal protective equipment and essential medicines; a warm-base domestic fill-and-finish vaccine manufacturing capability, backed by a standing agreement with an international body modelled on CEPI; a surge workforce roster of cross-trained and reserve clinicians with convertible ward space; and an enhanced surveillance and genomic-sequencing network feeding a rapid-response team. The unit's first move is to refuse the committee's framing. A conventional expected-value calculation — annual pandemic probability multiplied by expected loss, discounted, divided by portfolio cost — makes the portfolio look poor, because the annual probability is small and the expected annual loss looks modest. The unit shows this calculation deliberately, then explains why it is the wrong test: the loss is fat-tailed and correlated, and the country is risk-averse to ruin.

Instead, the unit models the tail. It builds three severity scenarios — moderate, severe, and catastrophic — and presents the tail losses directly: the catastrophic scenario's combined mortality, output, and emergency-borrowing cost exceeds the portfolio's lifetime cost many times over, even after applying a small probability. It values each portfolio component as a real option: the warm-base manufacturing capability is a call option on domestic doses months earlier than import queues would allow, and the unit estimates the value of those earlier doses in avoided deaths and avoided lockdown days using the value of a statistical life for mortality and scenario ranges for economic disruption. Surge capacity is defended on option grounds against the efficiency reviewers who want to cut idle beds: the slack is under-utilized by design and earns its return only in the tail. The stockpile is costed as a rotating, managed asset with an explicit replenishment schedule, not a one-off purchase.

The uncomfortable findings are two. First, the single most cost-effective component is the cheapest — surveillance and rapid response at the source — because early containment averts the tail rather than merely coping with it; the glamorous manufacturing plant is valuable but ranks below the unglamorous laboratory network. Second, the appraisal cannot honestly produce a single cost-per-QALY figure for the whole portfolio, because the largest losses are society-wide and non-marginal; the unit presents them as scenarios and is transparent that the ratio is not the decision rule. The recommendation that lands is about durability as much as content: fund the portfolio through a multi-year ring-fenced settlement and a standing statutory body with a legislated replenishment schedule, precisely because the appraisal predicts that an annual budget line will be cut in the neglect phase now beginning. The lesson generalizes — the scarcest resource in preparedness is not any single stockpile or plant but the institutional will to keep paying the premium after the fear has gone.

Four sector lenses

Startup

A small venture — a diagnostics developer, a surveillance-analytics firm, a platform-vaccine biotech — cannot survive on a market that only exists during emergencies. Its economics depend entirely on public market-shaping: advance purchase commitments, standing agreements with bodies such as CEPI, and warm-base subsidies that pay for readiness between events rather than only for product after an outbreak. The start-up's edge is speed and a novel platform; its exposure is the neglect phase, when preparedness funding dries up and the pipeline it built has no buyer. It should seek multi-year committed offtake and design its technology as a reusable platform — an option exercisable against many pathogens — rather than a bet on one.

Small business

An established small provider or supplier — a single laboratory, a regional personal-protective-equipment distributor, a community pharmacy chain, a GP partnership — is not chasing a novel platform but running a steady business that a public emergency will suddenly lean on. Its preparedness question is concrete and near-term: how much buffer stock, spare capacity, and supplier redundancy to carry when carrying it erodes thin margins in every normal year. Unlike the start-up, it cannot live on speculative public offtake; it needs preparedness to pay its way through standby and reserve contracts, rotation of stock into routine sales, and a place in a public replenishment schedule that turns idle buffer into a funded obligation. Its edge is being embedded, trusted, and locally quick to mobilize; its exposure is that a single large disruption can break it, so it should formalize its role in local response plans and price the slack it is asked to hold rather than absorb it silently.

Enterprise

A large enterprise — a multinational manufacturer, a hospital group, or a national insurer — is a holder of surge capacity and a counterparty to standby contracts. Its decisions on whether to maintain convertible facilities, reserve workforce, and diversified supply chains materially change a system's resilience, but each of those is idle capacity that ordinary efficiency metrics penalize. The enterprise must reconcile shareholder or board pressure for utilization with the option value of slack, and it is best placed to monetize that slack through paid standby and reserve contracts that turn preparedness into a revenue line rather than a cost. Enterprises with scale can also diversify supply geographically, converting a fragile just-in-time chain into a more resilient, deliberately redundant one.

Government

Government carries the accountability the others do not: it is the insurer of last resort, the funder of public-good capabilities markets will not supply, and the body that must ration ethically when capacity is overwhelmed. It optimizes for its population, yet its investment in surveillance and containment generates cross-border benefits — and its neglect, cross-border harms — as Chapter 4.2 — Global Health and Trade develops. Its hardest task is temporal and political: sustaining a preparedness settlement through the panic-and-neglect cycle, defending idle capacity against efficiency drives, and meeting standing obligations such as the International Health Regulations as a live capability rather than a filed form. Only government can credibly pre-commit the allocation rules, legislate the replenishment schedules, and pool the financing that makes national preparedness a durable option rather than a fading memory.

Common failure modes

• Appraising preparedness on expected cost per QALY. The ordinary ratio always makes a rare, correlated catastrophe look unaffordable per unit of expected benefit. Fix: frame as insurance and option value, model the tail, and state that the ratio is not the decision rule.
• Collapsing the loss to its mean. A single expected-annual-loss figure hides the fat tail that is the whole point. Fix: present severity scenarios and show tail losses directly.
• Treating a stockpile as a one-off purchase. Warehouses fill with expired or wrong-specification supplies and count as preparedness while being useless. Fix: manage a rotating portfolio with an explicit replenishment and shelf-life schedule.
• Letting efficiency drives eat surge capacity. Idle beds and cross-trained staff are under-utilized by design and are first to be cut. Fix: defend slack on option-value grounds, or buy the option through convertible facilities and reserve contracts.
• Building for the panic phase and ignoring the neglect phase. Capability bought at the peak is cut in the trough. Fix: protect it with multi-year ring-fencing, statute, standing contracts, and a standing body.
• Defending the border instead of investing at the source. Late, downstream response costs far more than early containment. Fix: fund surveillance, sequencing, and rapid-response teams as the front line.
• Deciding allocation rules during the crisis. Rationing improvised under pressure is unfair and slow. Fix: pre-agree, publish, and rehearse a procedurally fair framework.
• Using the precautionary principle as a blank cheque — or ignoring it entirely. Both extremes fail. Fix: act early under uncertainty for catastrophic threats, but bound the spend and prioritize no-regret measures.

Maturity model

Checklist

• Frame preparedness as insurance and option value, and state plainly that expected cost per QALY is not the decision rule.
• Model the loss distribution's tail with moderate, severe, and catastrophic scenarios; show tail losses directly rather than a single mean.
• Value each capability — stockpile, surge, manufacturing platform, pipeline — as a real option, and value the flexibility, not just the expected use.
• Prioritize upstream investment: surveillance, genomic sequencing, and rapid containment at the source before border defence.
• Manage stockpiles as a rotating portfolio with explicit shelf-life, replenishment, and carrying costs.
• Defend surge capacity on option-value grounds against efficiency drives; buy the option cheaply via convertible facilities and reserve contracts where full slack is unaffordable.
• Protect the commitment against the neglect phase with multi-year ring-fencing, statute, standing contracts, and a standing body.
• Use market-shaping instruments (advance market commitments, CEPI-style agreements, warm-base subsidies) to pull forward capacity markets will not supply.
• Count the whole social loss, monetize avoided deaths with the value of a statistical life, and be transparent about the correlated losses appraisal cannot fully quantify.
• Rehearse the response, measure capability decay, and meet International Health Regulations obligations as a live, exercised capacity.
• Pre-agree, publish, and rehearse a procedurally fair allocation framework for scarce resources before any crisis.
• Apply the precautionary principle with a bounded budget and no-regret measures, not as a blank cheque.

Key sources

• World Health Organization — International Health Regulations (2005) and core-capacity requirements; pandemic preparedness and response guidance.
• Coalition for Epidemic Preparedness Innovations (CEPI) — the 100 Days Mission for rapid countermeasure development.
• The Pandemic Fund (World Bank and WHO) — financing for pandemic prevention, preparedness, and response in low- and middle-income countries.
• HM Treasury — The Green Book: appraisal and evaluation, including guidance on risk, uncertainty, and optimism bias, and the value of a statistical life in appraisal.
• G20 High Level Independent Panel on financing the global commons for pandemic preparedness and response — the economic case for preparedness financing.
• Organisation for Economic Co-operation and Development (OECD) — analyses of the economics of preparedness and health-system resilience.

References

1. Pandemic prevention — Wikipedia — https://en.wikipedia.org/wiki/Pandemic_prevention
2. Tail risk — Wikipedia — https://en.wikipedia.org/wiki/Tail_risk
3. Expected value — Wikipedia — https://en.wikipedia.org/wiki/Expected_value
4. Expected utility hypothesis — Wikipedia — https://en.wikipedia.org/wiki/Expected_utility_hypothesis
5. Real options valuation — Wikipedia — https://en.wikipedia.org/wiki/Real_options_valuation
6. Option value (cost–benefit analysis) — Wikipedia — https://en.wikipedia.org/wiki/Option_value_(cost%E2%80%93benefit_analysis)
7. Surge capacity — Wikipedia — https://en.wikipedia.org/wiki/Surge_capacity
8. Value of life — Wikipedia — https://en.wikipedia.org/wiki/Value_of_life
9. Public good (economics) — Wikipedia — https://en.wikipedia.org/wiki/Public_good_(economics)
10. Precautionary principle — Wikipedia — https://en.wikipedia.org/wiki/Precautionary_principle
11. Global catastrophic risk — Wikipedia — https://en.wikipedia.org/wiki/Global_catastrophic_risk
12. Coalition for Epidemic Preparedness Innovations — Wikipedia — https://en.wikipedia.org/wiki/Coalition_for_Epidemic_Preparedness_Innovations
13. World Health Organization — International Health Regulations (2005) — https://www.who.int/health-topics/international-health-regulations
14. Coalition for Epidemic Preparedness Innovations — the 100 Days Mission — https://100days.cepi.net/
15. HM Treasury — The Green Book: appraisal and evaluation in central government — https://www.gov.uk/government/publications/the-green-book-appraisal-and-evaluation-in-central-government