A country has a profit and loss account, which is its balance of payments. It has a capital structure, which is its debt. It has a credit rating, which is the market's verdict on how it is run. It has a product and a market, which economists call comparative advantage. Read in that light, most of what states do to each other is ordinary commercial behaviour conducted with larger instruments.

There is one difference and it decides everything that follows. A company that loses is wound up and its people go elsewhere. A country that loses is still there afterwards, with the same population inside it. That single asymmetry is why a state can choose destruction and a firm cannot, and it is the reason the arithmetic below has two columns instead of one.

You do not need to be my friend to respect it, because you and I are going towards the same thing and we help each other.

That sentence is the whole of the cooperative case, and it is not sentimental. Nothing in the record suggests that states which cooperate do so because they have grown fond of one another. They cooperate when defection is detectable and expensive, and when the objective they share is worth more to each of them than the thing they would gain by breaking it. Friendship is not an input. Verification is.

A note on method

What follows is a constructed scenario, not a forecast, and no real country is named. The actors are archetypes defined by economic function, because function is what determines how a shock travels: the energy exporter, the chokepoint state that controls a strait, the fabricator that makes the advanced components, the granary that feeds other people's populations, the creditor that holds other people's debt, the neighbour that shares a border, and the distant country that believes it is uninvolved.

Every magnitude used below comes from a documented episode. When the text says that war-risk insurance rises from a fifth of a per cent of a ship's value to three per cent, that is what has actually been charged. When it says fertiliser prices rise by thirty-one per cent, that is a published forecast from a development bank. The scenarios are constructions; the elasticities inside them are not invented.

The channels

Before either column, the list of wires along which anything travels between countries. There are six and they are the entire apparatus.

Energy, which sets the price of everything else, because it is an input to every other input. Food, which is energy plus fertiliser plus freight, and which is the channel that reaches a household fastest. Shipping, which is how about four fifths of the physical trade of the world moves, and which is priced by insurers before it is priced by carriers. Money, which moves faster than anything else and therefore arrives first: sovereign spreads, currencies, capital flight, the cost at which a state can borrow. People, who move when the other five fail. And attention, which decides what any of this is understood to mean, and therefore what governments are able to do about it.

A country can be hit through any one of these without a single shot being fired at it.

PART ONE: COUNTRIES

1. WAR

Year zero: the first ninety days

The first thing that moves is not an army. It is an insurance premium.

Underwriters reprice a route within hours of a credible threat to it, and the repricing is brutal because the instrument is simple: a percentage of the value of the ship and its cargo, per voyage. On a route where war risk had been charged at a fifth of a per cent, premiums go to one per cent, and on the worst legs to three. In absolute terms a passage that cost between ten and twenty thousand dollars to insure costs between one hundred and fifty thousand and half a million. At that price the route is not dangerous. It is uneconomic, which is a different and more decisive thing.

So the ships go round. Transits through the affected passage fall by roughly a third. Voyages lengthen by ten to fourteen days, and rerouting adds somewhere between two and four hundred dollars to the cost of moving each container. Freight rates on the main affected lane spike by forty to sixty per cent, then settle twenty-five to thirty-five per cent above where they were. Nothing has been sunk. The map has simply been redrawn by an actuary.

In the same ninety days, three other things happen at once.

Energy reprices on expectation rather than on supply. Nothing needs to stop flowing; it is enough that the market believes it might. Gas and oil futures move first, electricity follows within the quarter in markets where power is priced off gas, and every industrial process that uses heat recalculates its margin.

Money leaves. Capital does not wait for clarity, and the first casualties are not the combatants but the countries that look adjacent on a map. Sovereign spreads widen for the entire neighbourhood, currencies of the region weaken against the dollar, and governments that were planning to issue debt discover that the window has closed. This is the fastest channel and the one that punishes the uninvolved most unfairly.

And fertiliser moves, which is the quiet one that matters most. Nitrogen fertiliser is made from natural gas, so a gas shock is a fertiliser shock with a delay of weeks. When a chokepoint carrying fertiliser and its feedstock was disrupted, urea rose forty-six per cent in a single month. Fertiliser prices are forecast to rise by about thirty-one per cent across 2026, to their least affordable level in four years.

Nobody outside the conflict has yet lost anything they can see. Every one of them has already paid.

Year one: the shock reaches the kitchen

The fertiliser price of year zero becomes the harvest of year one, and this is the mechanism by which a war between two states starves people on a different continent.

A farmer facing fertiliser at a third more than last year does not stop farming. He uses less. Yields fall by a few per cent across an entire region, which is invisible to him and enormous in aggregate. Grain prices rise, and because grain is the input to meat and to processed food, the rise arrives in every aisle of every shop. In countries where food is a tenth of household spending this is an irritation. In countries where it is half, it is a political event. The World Food Programme's estimate of what a single regional conflict can do, through this channel alone, is on the order of forty-five million additional people in acute hunger.

Meanwhile the belligerents themselves are discovering the arithmetic of modern war, which is that it is an industrial activity financed by debt. Defence spending moves from the two per cent of national output that was normal in Europe for a generation towards five, which several states on the eastern edge of the alliance already exceed. For European members alone, moving to three and a half per cent on defence is an additional three hundred and fifty billion dollars a year. Across the full alliance, a commitment to five per cent implies something in the order of two point seven trillion dollars of additional annual spending.

That money is not free and it is not new. It is borrowed, which raises the stock of debt, or it displaces something, which is almost always public investment, because investment is the only line in a budget that can be cut without anyone noticing within the electoral cycle. The thing being postponed is the thing that would have produced growth in year seven.

Years two and three: the second ring

The countries that share a border with the conflict have by now absorbed three separate shocks: the loss of a trading partner, the arrival of people, and the permanent repricing of their own credit.

Refugee movements are, in the medium term, fiscally positive and politically expensive, which is the worst possible combination for a government. The arithmetic is well established: the initial cost is immediate and concentrated in housing, schooling and health, while the benefit arrives through the labour market several years later and is dispersed. Governments that need to win an election inside two years will treat a net positive as a net negative, because the timing of a cost is as important as its size.

Trade reroutes permanently. This is the part that does not reverse when the shooting stops. A buyer who has spent eighteen months qualifying a new supplier in a different country does not switch back for a modest discount, because the cost of that switch was paid once and the relationship now has its own inertia. Market share lost during a conflict is lost for a decade.

The third ring, the countries that merely trade with the countries that trade with the belligerents, experiences the whole thing as a mystery. Their inflation is higher, their currency is weaker, their borrowing costs more and nothing has happened to them. This is the most under-reported fact in the economics of war: the overwhelming majority of the people who pay for it have no connection to it and no say in it.

Year five: the shape sets

By the fifth year the conflict has stopped being an event and become a structure.

Emergency spending has become a permanent line rather than a temporary one. This is the normal fate of any budget item created under pressure: it acquires suppliers, payrolls and constituencies, and a line with a payroll in every region is extremely difficult to reduce afterwards. It happens irrespective of anybody's intentions, and it is why the fiscal effect of a crisis outlasts the crisis by a generation.

Energy systems rearrange themselves around the assumption of insecurity, which is expensive but durable: duplicated routes, strategic reserves, long-term contracts signed at bad prices for the comfort of knowing the price. Security is purchased at the cost of efficiency, and once purchased it is rarely sold back.

And the debt taken on in years one to three begins to show up as an interest line that crowds out everything discretionary. A state that spends a growing share of its revenue on debt service and defence has, in practice, stopped being able to choose what it does.

Year ten: the bill

The visible costs of a war are the ones that can be photographed. The expensive ones cannot.

The largest single item is the investment that was never made: the decade of infrastructure, education and research that was displaced by security spending and debt service. It does not appear in any account because it consists of things that do not exist. The second is demographic, in the countries directly involved, and it operates over forty years rather than ten. The third is the fragmentation of the standards and institutions that made trade cheap, which raises the cost of everything by a few per cent forever, and a few per cent forever is a larger number than any war budget.

And the fourth is the one that sets up the next column: trust, which is not a sentiment in this context but an asset with a price. When it is present, a contract between strangers is enforceable at low cost. When it is absent, every transaction needs collateral, verification, insurance and a lawyer, and all four are charged to the buyer.

2. PEACE

Now run the same world with the same actors and the opposite decision, and note that nothing in this column requires anybody to become anybody's friend. Each step below is an arrangement that was adopted, historically, by parties who disliked each other intensely.

Year zero: the first ninety days

The first thing that moves is, again, an insurance premium, and it moves the other way.

Risk is priced, and a credible arrangement that makes a route safe is reflected within weeks: the war-risk component falls back towards a fifth of a per cent, voyages shorten by ten to fourteen days, and the two to four hundred dollars per container of rerouting cost disappears. Freight rates fall from twenty-five to thirty-five per cent above normal to normal. Everyone who buys anything that arrives on a ship receives a discount, and almost nobody notices, because a cost that stops being charged does not make the news.

Money arrives in the same order it left and with the same speed. Sovereign spreads narrow across the whole neighbourhood, not only in the countries directly involved, and a narrowing of a hundred basis points on the debt of a medium-sized state is a saving measured in billions a year, every year, available immediately and payable to no one.

Energy reprices downwards on the same expectation mechanism that moved it up: no new molecule is produced, the risk premium is simply withdrawn.

Year one: the mechanics of agreement without affection

This is where the piece has to be specific, because this is the part everyone assumes is impossible.

The first mechanism is the pooled input. If two states each depend on the other for something neither can replace, the cost of attacking includes the loss of that thing. The historical case is exact: two countries that had fought three wars in seventy years pooled the control of coal and steel in 1951, which were precisely the two industries required to build an army. The arrangement was not a declaration of friendship. It was a deliberate act of mutual hostage-taking, and it worked.

The second is the frozen dispute. A quarrel does not have to be won in order to be set aside. In 1959 the states claiming sovereignty over a continent agreed to suspend their claims without renouncing them, and to admit inspectors anywhere, at any time, without notice. The dispute still exists. It has simply been removed from the list of things worth fighting over.

The third is verification. Every agreement of this kind that has held had inspection inside it, and every one that lacked inspection failed. States that are not on speaking terms accept foreign auditors in their airports under the international civil aviation safety programme, admit inspectors to their stations on a whole continent under the 1959 treaty, and report their own compliance under the ozone protocol to a body that checks it. That is cooperation sustained by instruments rather than by goodwill, and instruments do not change their minds.

The fourth is conditional access. The most successful environmental treaty ever concluded worked because it had three features: a measurable target, trade restrictions against non-parties, and a fund to pay for compliance by countries that could not afford it. Non-participation was made more expensive than participation. No appeal to virtue was required at any point.

The fifth is the shared standard, and it is the one nobody mentions because it is already universal. States that point missiles at each other use the same internet protocols, the same container dimensions, the same air traffic rules, the same postal system and the same units of measurement. That cooperation is total, continuous and invisible, and it survives every crisis, because a standard requires no trust at all: it requires only that both parties prefer being understood to being misunderstood.

The common objective does not have to be noble. It has to be worth more to each party than what they would gain by defecting, and the defection has to be visible. Those are the only two conditions.

Years two and three: the dividend begins

Three things compound once risk falls, and they compound quietly.

The cost of capital falls across the entire region, not only in the countries that made the agreement. Every project that was marginal at the old discount rate becomes viable at the new one, and the number of such projects is very large, because investment decisions cluster just above and just below the threshold.

Trade grows faster than output, which is the normal condition when barriers and risk premiums fall, and trade growth is the mechanism that moves resources to where they are most productive without anybody planning it.

And insurance, freight, storage and inventory all fall together. Firms hold less stock when supply is reliable, and inventory is capital doing nothing. Releasing it is equivalent to a one-off transfer of several per cent of working capital into whatever the firm does next.

Year five: what gets built instead

The mirror of the war column's permanent defence sector is the peace column's permanent infrastructure.

The historical reference point is explicit. Between 1990 and 2008 military spending as a share of output fell by roughly half against its Cold War level; in Europe it went from about two and a half per cent of output in the early 1990s to about one and a half by 2014. During that period public debt fell in the countries that took the dividend, deficits turned into surpluses in some of them, and the capital released went into networks, education and the commercial internet.

That is what a peace dividend actually is: not a cheque, but a decade in which the most capable engineers in a country work on what that country will still need in thirty years, and in which the state can borrow for things that produce a return.

Year ten: compounding

By year ten the two columns have diverged by more than the sum of the annual differences, because the mechanisms are multiplicative rather than additive.

In the peace column, lower risk means lower capital cost, which means more investment, which means higher productivity, which means a larger base to tax, which means both lower borrowing and more spending capacity, which lowers risk further. In the war column the identical loop runs backwards.

Neither column is miraculous. They are the same loop with the sign reversed, and the sign is set in year zero by a decision that at the time looks like a matter of pride.

The same applies to businesses, and it is used all the time.

PART TWO: COMPANIES

The same two futures, read from a balance sheet.

A country can absorb a bad decade. A company cannot, which is why firms are the most sensitive instrument available for measuring what a political decision actually costs. They reprice within days, they report every quarter, and they are obliged to tell their owners the truth about it.

What follows looks only at ordinary productive companies: the ones that move things, insure them, grow them, process them, sell them and finance them. Each appears because its mechanics illustrate a channel, not because of any judgement about the firm itself.

1. WAR

The carriers

Shipping is where a geopolitical event becomes a line item first, because a vessel is a floating asset with a daily cost and a fixed route.

For a container operator of the scale of Maersk, Hapag-Lloyd or CMA CGM, the closure of a major passage is not primarily a danger problem. It is a scheduling problem with a compounding effect. Adding ten to fourteen days to a round trip removes those ships from the market for the duration, which tightens available capacity across the whole network and pushes rates up by twenty-five to thirty-five per cent on the affected lane. The carrier's revenue per box rises and its cost per box rises too, by two to four hundred dollars in fuel, crew and port time.

The result is a sector that appears to do well in the quarterly numbers while its customers are paying for a service that has got worse. Nobody gains anything that did not come out of somebody else's margin, and when the route reopens the capacity returns all at once and rates fall faster than they rose.

The second-order effect is the one that lasts. Every shipper that redesigned a supply chain around the longer route has written off that redesign cost already, and will not pay it again to go back quickly.

The underwriters

Marine war risk is where the price of a political event is set, and it is set before any government has finished its first statement.

The instrument is a percentage of the insured value, charged per voyage. On a route where the war-risk rate had been around a fifth of a per cent, it rises to one per cent and, on the worst legs, to three. In cash terms, a voyage that cost between ten and twenty thousand dollars to cover costs between one hundred and fifty thousand and half a million. The underwriters of the Lloyd's market and the mutual clubs that cover most of the world's tonnage are not predicting a war; they are pricing a probability distribution, and the premium is the clearest number available anywhere for what the market believes is about to happen.

For the insurer the business is not obviously attractive either. Premiums rise, but so does the probability of a total loss on a hull worth a hundred million dollars, and reinsurance costs rise behind it.

The airlines

Closed airspace does to an airline what a closed strait does to a carrier, with a worse cost structure, because fuel is a third of operating cost and an aircraft cannot slow down to save money the way a ship can.

Carriers whose geography forces them to overfly or circumnavigate the affected region add hours and fuel to every rotation; carriers that route elsewhere gain a structural cost advantage they did nothing to earn. This is the purest example in the whole piece of how war redistributes between parties with no stake in it: two competitors on the same route, with the same aircraft, suddenly have materially different unit costs because of a line on a map.

The energy companies

Shell, TotalEnergies, Equinor and their peers are price-takers, which is the point. They do not set the risk premium; the market does, and their revenue moves with it while much of their cost base does not.

The consequence runs the other way for everyone downstream. Every manufacturer that melts, heats, dries or moves anything recalculates its margin within the quarter, and the ones with the thinnest margins and the least pricing power, which is most of the small industrial base of any country, are the ones that go first.

The fertiliser and food chain

This is the channel with the longest fuse and the widest blast radius.

Nitrogen fertiliser is made from natural gas, so a gas price shock arrives at Yara, Nutrien and Mosaic as a cost shock within weeks. When a chokepoint carrying fertiliser and its feedstock was disrupted, urea rose forty-six per cent in a single month, and fertiliser prices are forecast to rise about thirty-one per cent across 2026, to their least affordable level since 2022.

The farmer does not stop farming. He applies less, and the consequence appears a season later as a yield a few percentage points lower across an entire region. That arrives at Nestlé, Danone, Unilever and every other processor as a raw material cost, and at the household as the price of bread. The distance between a political decision and a kitchen is about four months, and it runs through a fertiliser plant.

The manufacturers and the retailers

Any company that holds little inventory on purpose discovers what that policy costs when it is tested. Just-in-time manufacturing, as practised in the automotive sector by Toyota, Volkswagen and everyone who copied them, is an optimisation that assumes the sea stays open.

The counter-example is instructive, and it is a retailer. Inditex built its model on short lead times and nearby production specifically so that it could change what it makes inside a season. That design, which was adopted for commercial reasons entirely unconnected to geopolitics, happens to be the single best defence against a freight shock that exists. The firms that spent thirty years moving production as far away as possible in search of unit cost are the ones now paying to move part of it back, and that reversal is a capital expenditure with no new output attached to it.

The banks

Trade finance is the invisible plumbing of about eighty per cent of world trade, and it reprices with the risk.

HSBC, Santander, Standard Chartered and the rest of the trade banks widen their pricing on letters of credit, shorten tenors and reduce limits for counterparties in the affected region. For an exporter in a neighbouring country that has done nothing, the practical effect is that the credit which made the shipment possible costs more or is not available. Sovereign spreads widen across the neighbourhood, and because banks hold their own government's debt, a widening spread is simultaneously a funding cost and a balance sheet loss.

The platforms

Advertising is the first budget line any company cuts when it cannot see six months ahead, which makes Meta, Alphabet and Amazon unusually sensitive instruments for consumer confidence.

In an escalation, their revenue does not collapse, but its growth rate falls, and it falls first in the categories closest to discretionary consumption: travel, retail, entertainment. The structure described elsewhere in this newspaper holds, with three intermediaries taking between sixty-two and sixty-four per cent of around $740 billion of digital advertising this year, but the pot itself stops growing. The same auction with less money in it.

The semiconductor chain

And underneath all of it sits the concentration already documented in these pages: around ninety per cent of the world's most advanced chips are manufactured on a single island, with close to ninety-nine per cent of the chips used to train frontier artificial intelligence models originating there, and the advanced packaging capacity being duplicated elsewhere does not arrive until the end of the decade.

TSMC, ASML and the handful of firms around them are not a sector in this scenario. They are a single point of failure for every other sector listed above, because there is no modern product without them. A disruption here is not a cost increase. It is a stop.

2. PEACE

The carriers

Routes reopen, voyages shorten by ten to fourteen days, the two to four hundred dollars per container of rerouting cost disappears and freight rates fall from twenty-five to thirty-five per cent above normal back to normal. Capacity floods back into the market all at once, which is excellent for every company that buys shipping and difficult for the companies that sell it, since the ships that were absorbed by longer routes return looking for cargo.

The gain is diffuse and the loss is concentrated, which is why the lobbying is always louder on one side than the other.

The underwriters

War-risk rates fall back towards a fifth of a per cent and the premium disappears from the freight bill. Nobody writes about it. A cost that stops being charged is the most invisible form of prosperity there is.

The airlines

Airspace reopens and routes shorten. For a long-haul carrier, removing an hour of flight from a daily rotation is worth more over a year than almost any commercial initiative available to it, and it arrives without a single decision being taken inside the company.

Energy and its customers

The risk premium is withdrawn and the price falls without a single additional molecule being produced. Every industrial process in every importing country improves its margin simultaneously, which is the broadest and least visible form of stimulus that exists, because it reaches every firm at once without passing through a government.

Fertiliser, food and the four-month fuse

The fuse runs in both directions. Cheaper gas is cheaper fertiliser within weeks, normal application rates within a season, normal yields a season after that, and the price of bread stops being a political question within about a year.

This is the single highest-return channel in the entire peace column, measured in human terms rather than financial ones, because the people at the end of it have no other margin to give.

The manufacturers and the retailers

With reliable supply, firms hold less inventory, and inventory is capital doing nothing. Releasing it transfers several per cent of working capital into whatever the company does next, across every manufacturer and retailer simultaneously.

More importantly, the capital expenditure that was going to be spent on duplicating supply chains for safety is not spent on duplication. It is spent on capacity, which produces additional output, which is the difference between an economy that is busy and an economy that is growing.

The banks

Spreads narrow, tenors lengthen, trade finance limits reopen and the cost of capital falls across the whole region. A hundred basis points off the cost of borrowing is a saving that applies to every project in every company in every year, and it tips a very large number of marginal investments into viability, because investment decisions cluster around the threshold.

This is the quietest and largest number in the peace column, and no government can claim credit for it.

The platforms and the consumer economy

Confidence returns before the data does. Advertising budgets are restored early in the cycle because they are the cheapest commitment to reverse, so the platforms recover before the firms advertising on them, which is a reliable and slightly uncomfortable regularity of the modern economy.

The semiconductor chain

The single point of failure remains a single point of failure. Peace does not fix concentration; it only postpones the consequences of it, and the duplication of capacity in other geographies proceeds on its own timetable, which is measured in years and is not affected by anybody's goodwill.

That is the one line that is identical in both columns, and it is worth noticing.

What the two columns have in common

In both futures the same four facts hold.

The costs and the benefits are paid by people who are not consulted. The fastest channel is money and the slowest is demographics, so the political reaction always arrives calibrated to the wrong one. The cheapest interventions, the ones that cost almost nothing and return the most, are the boring technical arrangements: an open route, a predictable rule, a verified commitment. And the single point of failure in the supply chain of the modern world is unaffected by either column, which means the one risk nobody is arguing about is the one that would stop everything.

THE CONCLUSION

What they should do is simple: common objectives. You do not need to be my friend to respect it, because you and I are going towards the same thing and we help each other. And how they can come to set themselves a common objective of mutual growth, choosing a path different from destruction, through cooperation out of pure ambition.

That is the case, and the two columns above are what it is worth in money, in food and in years.

Now the objection, which has to be stated as plainly as the case.

The normal thing is for distrust to make the obtaining of resources turn one stronger than the other, so it is probable that this remains a utopia, because someone powerful can act left and right, and then the same ambition that causes cooperation separates it in parts, this being quite a dense and complex subject.

That is the flaw in the mechanism and no arrangement in this piece repairs it. Every instrument described in the peace column, the pooled input, the frozen dispute, the inspector, the conditional access, the shared standard, works by making defection visible and expensive. None of them works when one party becomes strong enough that the cost of defecting is lower than the gain. Ambition builds the agreement and ambition dissolves it, and the interval between those two moments is what we call an era of peace.

Although in the end wars were there probably from before humanity, however, it is also part of real evolution to overcome it in order to live in harmony and ascend towards what we deserve.

Self-destruction or unlimited freedom, two paths for humanity, or partial destruction being the third path, which is the one usually chosen.

The third is chosen because it is the only one that does not require anybody to decide anything. The first two demand a resolution; partial destruction is what happens when nobody resolves. It is the default setting of the species, it has run for ten thousand years, and the columns above are what it costs per decade.

What this piece can establish, and the limit of what it can establish, is this. The arrangements that produce the right-hand column exist, they have been built before by parties who detested one another, and they are cheap. Whether they are adopted is not an economic question and nothing in a balance sheet will answer it.

Sources

Shipping, insurance and freight. War-risk premium levels, the rise from around 0.4 per cent to 1 per cent and up to 3 per cent of hull value, the shift from $10,000-20,000 to $150,000-500,000 per voyage, the 25-35 per cent elevation of Asia-Europe freight rates, the 10-14 additional transit days, the $200-400 per container rerouting cost and the fall of roughly 30 per cent in transits through the affected passage are drawn from marine insurance market reporting and shipping trade coverage during the disruption of the Red Sea route.

Defence budgets. Figures for alliance spending, the 2 per cent threshold being met by all members, national shares above 5 per cent of output, and the commitment agreed in 2025 to reach 5 per cent by 2035, are published by NATO. The estimates of $350 billion of additional annual spending for European members at 3.5 per cent, and of roughly $2.7 trillion across the alliance at 5 per cent, are published analyses of those commitments.

The peace dividend. The halving of military spending as a share of output between 1990 and 2008, the fall in European defence spending from about 2.5 per cent of output in the early 1990s to about 1.5 per cent by 2014, and the associated fiscal effects, are documented in International Monetary Fund and central bank research of the period.

Food and fertiliser. The 46 per cent month-on-month rise in urea prices following disruption to a chokepoint, the forecast 31 per cent average increase in fertiliser prices in 2026 and the assessment of least affordable levels since 2022 are from the World Bank's Commodity Markets Outlook. The estimate of 45 million additional people in acute hunger is from the World Food Programme.

Company figures. All company references are to disclosed operating characteristics and publicly reported figures; no projection is made about the future results of any named firm.

Historical arrangements. The pooling of coal and steel production in 1951, the suspension of territorial claims and the inspection regime agreed in 1959, the inspection provisions of the international civil aviation safety oversight programme, and the structure of the 1987 protocol on substances that deplete the ozone layer, including its trade provisions and its multilateral fund, are matters of treaty record.

Scenarios. The two columns are constructed scenarios, not forecasts. No country is named in them. The magnitudes used are drawn from the documented episodes above and applied to archetypes defined by economic function.