However, the spread of monkeypox has been very different to the spread of Covid. While Covid is spread through the air, by droplets on patients’ breath, monkeypox is mainly spread by skin-to-skin contact. And because the initial outbreak arose among gay and bisexual men — many of the initial cases can be traced back to a Pride event in the Canary Islands, a sauna in Madrid, and other venues — the large majority of cases have, so far, been among men who have sex with men (MSM).

That said, the disease is not specific to gay men, or sexually transmitted in the strict sense of the word, and could reach the wider community. Even if it doesn’t, it’s important to get a sense of how far monkeypox could spread among MSM, and to try to minimise its damage and understand the scale of the public health problem. With that in mind, we gathered ten forecasters for a discussion of the likely outcomes of the monkeypox outbreak — to try to predict both the likely size of the outbreak, and how it will look: will it break out of the MSM community in large numbers?

A note on forecasting methods:

The classic “superforecasting” model requires forecasters to give a percentage probability for a certain outcome, such as “Britain will vote to remain in the EU, 66%”. Forecasting skill is graded on how well-calibrated those forecasts are — do your 66% forecasts come in 66% of the time? — and also how confident they are: getting a 90%-confident forecast right is rewarded more highly than getting a 55%-confident forecast right. (On the other hand, getting a 90%-confident forecast wrong is punished more severely than getting a 55%-confident forecast wrong.)

In this case, we do something similar, but, in a way, reversed. The forecasters gave a range of values that they thought was 80% likely to contain the true outcome. So, for instance, if one thought that it was 90% likely that the number of monkeypox cases in the UK before the end of 2022 would be higher than 500, and 90% likely that it would be lower than 10,000, their 80% range is 500 to 10,000.

Suggested reading

Superforecasting the end of Covid

By Michael Story

They also each gave a 50% value, which you can most easily think of as their central, most likely estimate. The forecasters will be graded on whether their 80% forecasts come true 80% of the time, as usual, but instead of being rewarded for making higher-confidence predictions, they’ll be rewarded for giving narrower ranges. For each question, we also gave a median of the forecasters’ 10%, 50% and 90% predictions — that is, if you took all 11 predictions and put them in a row from smallest to largest, we’d take the one in the middle.

The questions:

How many cases of monkeypox will there have been worldwide by the end of 2022?

• Median 10% estimate: 13,500
• Median 50% estimate: 45,000
• Median 90% estimate: 525,000

As of 17 June, there have been 2,103 laboratory-confirmed cases of monkeypox reported to the WHO. The forecasters think that even in a realistic best-case scenario, that number will only be a fraction of the eventual total: the median 10% (that is, low-end) estimate is six times higher. On the other hand, the median upper-end estimate, while a large number — more than half a million — is still negligible compared to the spread of Covid: by the end of 2020, there had been more than 80 million confirmed cases of Covid-19, and many times that number of undetected ones.

Yet it’s important to note that the forecasters’ predictions varied widely. Three of them pointed out that while the virus spreads through skin contact now and has a relatively low R number, it is mutating, and could easily become transmissible. One said: “Multiple somewhat distinct strains have already evolved, which leads me to be concerned that there is a fairly high amount of genetic drift occurring, which could lead to the situation turning into a more detrimental pandemic.”

Another said, citing this paper, that the virus seems to have a very high R within MSM communities, but an R below 1 outside it. (The R value is the number of people the average infected person goes on to infect.) But, they said, “as it spreads more the chance that it mutates to gain a transmissibility advantage increases, quite possibly via the airways”: as a result, that forecaster gave a 90th-percentile forecast of 30 million infections, the highest in the group by a factor of four.

Suggested reading

Inside the Covid ward

By Jane Smith

Other forecasters were more optimistic. One thought that the total would be no higher than 80,000; another, 200,000. Another with a relatively low estimate (median 40,000, 90th percentile 400,000) pointed out that “there has been limited transmission beyond networks of close contacts, so far”. Another, who had a very wide range (from 12,000 to 5 million) said: “R is clearly lower than for Covid, so there clearly is a plausible future where once governments get serious about mitigation it ceases to spread quickly. On the other hand, well, they might not.”

It’s worth noting that these estimates are of the total cases, not just the confirmed ones. (One forecaster said they “expect much of the world is vastly undercounting”.) This means it will be impossible to accurately grade who has got this question right.

How many cases of monkeypox will there have been in the UK by the end of 2022?

• Median 10% estimate: 6,000
• Median 50% estimate: 22,000
• Median 90% estimate: 85,000

As with the worldwide estimates, there was considerable spread. One — the same person who gave a high-end estimate of 30 million in the first question — used the same reasoning to say numbers could get as high as one million in the UK.

On the whole, though, the feeling among the forecasters was that the UK is better placed to control spread than most countries are. “UK monitoring is on the better end,” said one, “and there are initial attempts to vaccinate.”

But there were concerns. “It’s failed to be controlled particularly well so far, despite good monitoring,” one wrote. “It’s on the worse end of what experts would have hoped for a month ago, perhaps because of long incubation periods. There’s still a lot of opportunity for superspreader events, especially given Pride, and relatively little cautiousness so far. But awareness within LGBT communities should slow it down.”

As with the “worldwide” question, the forecasters were talking about the true number of cases, not confirmed cases. However, the monitoring in the UK is likely to be pretty comprehensive, so the two numbers should be comparable.

What percentage of 2022’s worldwide cases will be in men?

• Median 10% estimate: 50
• Median 50% estimate: 94
• Median 90% estimate: 99

At the moment, more than 99% of recorded monkeypox cases worldwide have been among men, suggesting that the spread has been almost exclusively among MSM. The forecasters, on the whole, expected that to remain the same, although they felt that there was a distinct chance, above 10%, that it could spread so rapidly among the general population that it could wash out the male head start altogether. One pointed out that once a disease gets outside the subpopulation in which it started, it can behave very differently depending on the context: “The US currently has a 5:1 male/female ratio of HIV,” wrote one, “while Sub-Saharan Africa has more women with AIDS than men. The modes of transmission do not keep the straight community safe. There are far more straight people than gay people, so once it starts to increase in that community, the potential population is much greater.”

Another forecaster thought that this is essentially the same question as the case-numbers question, somewhat repackaged. If it starts to spread very rapidly, by some other method than skin-to-skin contact, then it will spread outside MSM communities: “I think this is essentially the same as asking the chances it mutates to become more transmissible and spread outside of the networks it’s currently spreading through. It’s very hard for me to estimate this risk, but I think it increases as the case numbers increase.”

All 11 forecasters thought there was at least a 10% chance that cases among women would reach close to 50% of the total, but most thought that that was unlikely: only one put their central estimate at below 90% male, and four had it at 95% or higher.

What percentage of 2022’s UK cases will be in men?

• Median 10% estimate: 55
• Median 50% estimate: 98
• Median 90% estimate: 99

The forecasters felt the story was likely to be largely similar in the UK to worldwide, with some caveats. First, they felt the disease was likely to be better controlled here and to spread more slowly, so the chance of it spreading widely outside MSM communities was smaller, hence the higher median 10% estimate. Second, one noted that the UK’s HIV population has a higher male-to-female ratio than the global average, and used that to extrapolate to monkeypox.

Overall, the forecasters’ attitude can be summed up in this comment: “I think community spread is definitely possible, but founder effects and (reasonably) slow spread will mean the MSM community will still dominate the numbers by the end of the year.” One added that the UK government’s current strategy, of vaccinating high-risk MSM, is very good, and “may solve the problem on its own”. However, they added that public messaging at the moment is not especially useful, since it doesn’t warn at-risk communities, tending to emphasise the risk to wider society rather than to MSM groups. Changing that would, they said, be worthwhile.

Conclusions:

The forecaster group, on the whole, felt that monkeypox will continue spreading, both in the UK and worldwide. However, unless it mutates significantly to become capable of spreading in ways other than through skin contact, they felt that its numbers were likely to remain comparatively low, and that it was most likely to stay chiefly in the MSM community.

Yet they also felt there was a significant tail risk of a much wider impact. They noted that a novel disease which appears to be mutating relatively rapidly has a non-negligible chance of achieving new means of spreading — and that there was a greater than 10% chance of widespread transmission among the wider population, with hundreds of thousands or even millions of cases not unlikely.

***

A version of this research first appeared on the Swift Centre.

That’s not necessarily because they’re wrong. It’s because they are imprecise: the pundits won’t fully commit to anything. “The year ahead will be difficult for Boris Johnson,” that sort of thing. Whatever happens, it’ll be hard to say whether that prediction came true or not.

What’s much more worthwhile is making clear, well-defined, falsifiable forecasts, with fixed timetables: forecasts that commit you to say “OK, I got that wrong.”

I had a go here, here and here during the pandemic (an embarrassing number of which have been wrong). But since it’s the new year, I thought I might stake my own reputation on a few forecasts for 2022.

To do so, I’m going to use the Phil Tetlock “superforecasting” model, of giving lots of forecasts with levels of confidence. So I might say it’s 60% likely that Boris Johnson will have another child in 2022, or that there’s only a 10% chance that Keir Starmer will be prime minister by the end of the year. At the end of the year, I’ll look at all my forecasts. Ideally, about 60% of my 60%-confident forecasts should have come true, and about 10% of my 10%-confident forecasts, and so on[1.  Ideally ideally, I’d have made hundreds of forecasts and you could get a really good sample to see how well-calibrated I am. But unfortunately that’s not going to happen, because calculating these takes ages.].

I’ve used the prediction site Metaculus for my questions. When the site had a good question on it already, I’ve used that; when I needed to make up my own, I’ve done that. I’ve linked to the questions so you can make your own predictions and see what the wider Metaculus community thinks.

Coronavirus

I think that this year will see things return pretty much to normal in Britain. Every so often there’ll be a bit of a flare-up, and some people will call for an immediate return to mandatory masks and closed pubs and all that. But 90% of the population aged over 12 has now had at least one jab and 60% have had three, and most people have probably had the actual disease as well (13 million confirmed cases, which is likely an undercount by at least 60%). So the actual risk to British people is low.

That said, about 100 people are still dying every day with Covid. That sounds bad, and it is, but it’s only about one in every 16 deaths in the UK. Still, if it carries on at that rate for the whole year, we’d expect another 35,000 to 40,000 deaths this year. It may not carry on at that rate all year, but I can’t imagine it’ll drop so much that we see fewer than 20,000. So I’d say it’s 75% likely that we see 20,000 or more deaths within 28 days of a positive Covid test in Britain in 2022.

Also, I think we’ll get a few scares from new variants. The WHO has announced five “variants of concern” since the pandemic began, the most recent being Omicron. Given a base rate of about 2.5 a year so far, it’s pretty likely that we’ll see more this year. Whether any of them will be as devastating as Delta or Alpha is a bigger question, but I’d say it’s 90% likely that the WHO will announce at least one new variant of concern in 2022.

Relatedly, if new VoCs do arise, will one of them overtake Omicron as the most commonly detected strain of Covid in the world? Again, I think this is pretty likely – I’d say there’s only a 30% chance that Omicron will be dominant by next January. Six weeks ago we’d never heard of Omicron and Delta was dominant; the original Covid was overtaken by Alpha only eight months or so after it was declared a pandemic. So far no variant has remained dominant for more than a few months and Omicron appears to be petering out in South Africa.

Predictions:

More than 20,000 people will die within 28 days of a positive Covid test in the UK in 2022 (75%)

The WHO will add another variant to their Variants of Concern in 2022 (90%)

Omicron will be the globally dominant strain of Covid at the end of 2022 (30%)

Suggested reading

Could you predict Brexit?

By Tom Chivers

Vaccinations

The UK’s booster campaign has rolled out with impressive speed since it was announced in September. But there have been suggestions that we will need to do annual Covid jabs, as we do for flu. So I predict it is likely (65%) that we will offer fourth jabs to healthy British adults before the end of 2022.

Not that I think we should do this. I think the vaccines we’ve already had will probably protect people very effectively against severe disease and death even once the measurable antibodies have declined, and I think each new jab will do lots more good in the arms of unvaccinated people in lower-income countries than they will in the arm of a triple-jabbed pensioner in Reading. But given that governments have a duty to look after their own citizens, and, more pressingly, a need to curry favour with those citizens to avoid being voted out, I think it’s pretty likely that we’ll do it. Metaculus forecasters think it’s likely in the US, as well.

The important thing will be whether we vaccinate the world, and especially the poorest countries. It should be said that despite a widespread and understandable feeling that the West has hoarded all the vaccines, almost 60% of the world’s population has received at least one jab, and even lower-middle income countries (which include places like Bangladesh and Egypt) have vaccinated more than 50%.

But the truly poor countries – places like Afghanistan, Sudan, North Korea and many sub-Saharan nations – are lagging badly behind, at only about 8.5% coverage. It’s been speeding up, and in richer countries vaccinations have followed an S-shaped curve – slowly, then quicker and quicker, then slowing down again. But looking at the curve on the Our World in Data page, it looks like the low-income countries are taking off at about a quarter the speed that the rich ones did.

It took rich nations about six months to get to 50% coverage: the world makes vaccines faster now, but also there are still huge logistical and incentive problems for getting those vaccines to poorer countries. I’d guess that between 30% and 60% of inhabitants of the world’s poorest countries will have had at least one jab by 2022. I think it’s also pretty likely (75%) that at least one country will still have fewer than 10% vaccine coverage, just because there are a lot of poor countries, some of which have basically no working health infrastructure.

Predictions:

The UK will authorise fourth doses of Covid vaccines for healthy adults in 2022 (65%)

What percentage of people in low-income countries will have had at least one dose of a vaccine by 2023? – between 30% and 60%

At least one nation will have less than 10% of its population vaccinated with at least one dose by 2023 (75%)

Suggested reading

Can we trust the climate scientists?

By Tom Chivers

Climate

The world will continue to warm, on average, over the next few years. There’s too much carbon in the atmosphere already for that to change.

But there’s a lot of variability from year to year. Most years are somewhat hotter than the one before, but not always: of the 20 years leading up to 2020, 13 have been hotter than the previous year, and seven have been cooler, according to the Met Office. So the base rate is that it’s about 65% likely that 2022 will be warmer than 2021.

That said, 2021 has been a relatively cool year, due to La Niña, an atmospheric and oceanic phenomenon that brings cold water to the surface of the oceans. La Niña finished some months ago, so it shouldn’t affect 2022. I think it’s therefore very likely (85%) that 2022 will be hotter than 2021, when the data eventually comes in.

Prediction:

The global average surface temperature in 2022 will be higher than it was in 2021 (85%)

Suggested reading

The men who failed Britain

By Tom Chivers

Energy

There’s a cruel joke that goes to the effect that “fusion energy is 30 years away and always will be”. That said, in the last few years, there’s been a lot more excitement about it, and several new companies have started up, with nearly two billion dollars of funding backing them.

It’s very unlikely that fusion energy will reach any of its milestones – “ignition” or “breakeven” or profitability or anything – this year, and pretty unlikely that they’ll do it this decade. But I do think that the excitement will carry on building. According to a survey by the Fusion Industry Association, there are “at least 35” fusion companies around the world, most of which have been founded in the last 10 years. I think it’s likely (65%) that number will go up in 2022, as the technical questions continue to be answered, and as investors and entrepreneurs get more excited about the possibility of a viable, profitable industry.

Also, the cost of solar energy has plummeted over the past decades. It will almost certainly continue to do so, but since it is already cheaper than any form of fossil fuel, cost is no longer the sticking point. Now, the bottleneck for renewable energy is making it available when the sun isn’t shining or the wind isn’t blowing – having loads of sunshine in August is no good for heating your house in December.

There are various ways of doing that, such as installing cables between countries, so the UK’s offshore wind can supply Spain when it’s windy, and Spain’s solar can supply the UK when it’s sunny. But the most obvious is storage, and the most obvious form of electricity storage is batteries.

The cost of electricity storage has also dropped enormously – according to Bloomberg, it’s gone from a global average of about $300 per megawatt-hour in 2018 to $138 in 2021, mainly driven by the electric car industry. Despite supply chain issues and increased cost of various raw materials, mainly metals, I think that’s probably going to carry on – I think it’s 80% likely that the next Bloomberg report will find that the global average drops in 2022.

We’ll still rely on fossil fuels for some years to come. But I think in 2022 we’ll see some of the pieces of the zero-carbon future start to come together.

Predictions:

There will be more fusion companies at the end of 2022 than there were at the end of 2021 (65%)
The cost of battery storage will be lower per megawatt-hour at the end of 2022 than it was at the end of 2021(80%)

Suggested reading

Have we seen the end of Covid?

By Tom Chivers

Medicine

The biggest thing in medicine this year has been the Covid vaccines. And they’re pretty astonishing things. Remember that at the start of 2020, there were zero approved mRNA vaccines in the world: now billions of people have been given one, and millions of lives have been saved.

You can use mRNA vaccines for pretty much any disease, and in the next few years we’ll doubtless see them for all sorts of things. One that really interests me is cancer. Cancer “vaccines” aren’t like vaccines for measles or Covid – they’re treatments, given to people who have already been diagnosed with the disease, rather than prophylactics given to people at risk of contracting them. As I understand it, it involves presenting the immune system with the RNA sequence for a protein specific to that patient’s cancer, which will help the body fight that cancer. It’s a sort of personalised advance on existing immunotherapy, which has already turned various previously fatal cancers into largely manageable chronic illnesses. So this could be a really big deal in coming years.

Lots of drug companies have plans to make cancer mRNA vaccines. For instance, BioNTech has four ongoing Phase II trials. I think it’s pretty likely that we’ll hear good news from at least one of them, and that a Phase III trial with thousands of participants will be announced. Let’s say 55%.

Prediction:

A Phase III trial of at least one of BioNTech’s cancer mRNA vaccines will be announced in 2022(55%).

Suggested reading

What Warhammer taught me about football

By Tom Chivers

Football

Until a week or so ago Liverpool were one point off the lead and looking good: then they drew with Spurs, lost bizarrely to Leicester after missing a penalty, and then drawing with Chelsea despite being two-nil up after half an hour. And Man City should have lost to Arsenal but Arsenal decided to be Arsenal and gave away a penalty and had a man sent off within 10 minutes and threw it away.

So now Liverpool are 11 points off the top (albeit with a game in hand) and let’s face it the title race is over in January. I reckon only about a 10% chance of turning it around. Still, there’s always the Champions League.

Prediction: Liverpool will win the league this year (10%)

Conclusion

I wanted to add another couple of questions, but they’re hard to do good data on. For instance, I bet that global extreme poverty will be lower in 2022 than it was in 2021, as it has been almost every year for the last 40. But we won’t get good data for a few years. I also bet that even though the world will get better on most metrics, such as life expectancy, wealth, etc, an increasing share of people will still think that it’s getting worse, because that’s how we work. But that’s hard to get into a straightforward question, and again, even if we could, the data for 2022 won’t be around until 2025 or something anyway. Still, I think those things are probably true.

My final prediction is that these predictions will not be very accurate. I’d love it if you would have a go and see if you can do better.

Since then, while there have been proxy wars (notably Vietnam) and local conflicts, no big, powerful countries have gone to war with one another. But there are still tensions, and if you were betting on a war between two great powers now, China and the US would again be the most likely.

The Korean peninsula is probably not where you’d bet on it starting, though. Instead, you might look to Taiwan, the self-governing island off the south-east coast of China. Just this week, a Chinese magazine published a video showing a simulation of a ballistic missile attack on Taiwan, disabling its defences ahead of an invasion. A week before, China sent 28 warplanes, including nuclear-capable bombers, into the air defence zone around Taiwan. At the same time, the Chinese leader Xi Jinping gave a speech promising to “utterly defeat any attempt toward ‘Taiwan independence’.”

The US is an ally of Taiwan’s; it would be expected to defend it, should China attempt to invade. But just as in Korea, two-thirds of a century ago, the question is: would it? And will China invade?

Recently, a group of anonymous superforecasters have tried to answer that question, and shared their work with me. They previously had a go at forecasting the outcomes of the rise of various Covid variants; now, they are interested in whether we ought to worry about Chinese aggression over Taiwan.

Most people are rubbish at actual, falsifiable forecasts. If you ask questions like “Will there be a conflict between China and Taiwan in which at least 100 people are killed before 2026?” or “Will the US defend Taiwan in the event of a conflict between Taiwan and China?”, and ask them to give percentage likelihoods of each eventuality (60%, 20%, 99%), then most people do no better than random guessing. But some people do a bit better, and superforecasters do exceptionally well.

My six forecasters who were considering what would happen with Taiwan were asking several questions, but the key one is probably the one I mentioned above: how likely is it that we will see a conflict between China and Taiwan in which at least 100 people are killed?

The first thing they do is to look at base rates. This is also known as taking the “outside view” on a problem, as opposed to the “inside view”. In a classic example: you’re at a wedding and you’re asked whether the marriage will be a lifelong one. If you look at the couple and think “they look very much in love” and base your answer on that, then you’re taking the inside view.

If, instead, you look at how many marriages in general end in divorce – about 40%, in the UK – and start from there, you’re using the outside view. You then use the inside view to adjust from there: if you know that the couple are well-suited to each other, you adjust your estimate up; if you think they’re not, you might adjust it down.

Regarding Taiwan, one wrote: “An elementary application of base rate thinking means that we have to put the probability of China going for Taiwan within the next five years at below 50 percent.” Since 1949, when the nationalist government was defeated in the Chinese civil war and its leaders fled to Taiwan, mainland China has never invaded the island. If the odds of an invasion in any given five-year period were above 50%, then the chance of making it to 70 years without one would be about one in 16,000.

But, then, they take inside-view considerations into account. For instance: China is catching up on the US in terms of military might. It “is serious about taking control of Taiwan”, which it sees as a threat, and is making more aggressive noises. 

Taiwan itself is very badly defended and the defence it does have is ill-suited to the task (its military has “pursued a suicidal procurement strategy of expensive boutique US kit that will be no use in the crisis, like fighter jets that will be killed on the ground by the opening Chinese missile barrage”). The US military is aimed at fighting the War on Terror, not defending overseas territories against invasion. US public opinion might not support shedding blood over defending Taiwan. 

Suggested reading

Did anyone predict coronavirus?

By Tom Chivers

On the other hand: a war would be a huge risk for China; “every Chinese leader has an incentive to leave such a risky endeavour to his successor,” another forecaster writes. In the short term, the balance of power is still with the Americans, and China can afford to be patient and wait until the middle of the century. The forecasters use facts like these to adjust their initial base-rate estimate. 

The six forecasters estimate the likelihood of a significant China-Taiwan conflict at between 8% and 23% in the next five years, with a median estimate of 14%. That doesn’t sound all that bad, but it’s worth adding something.

If we’ve learnt anything from Covid, it should be that preparing for the most likely outcome is not enough. The odds of a global pandemic in any given year is probably only about 1%. But if one happens, it turns out, it’s really bad, and it would have been worth investing a significant amount of resources to avoid or mitigate it. One of the superforecasters told me that  “a 14% chance of a proper conflict by 2026 is quite a big deal. If someone says there’s a 10% chance of a really bad outcome, the expected value [the impact multiplied by the probability] is still really bad.” So you might not think a particular bad outcome is very likely, but if it’s bad enough, then you ought to prepare for it anyway.

In the case of Covid, it’s understandable that the UK government didn’t foresee in February that a global pandemic was coming in March. Lots of people were saying it was, others were saying it wasn’t. But they should have foreseen that a global pandemic was highly plausible, that there was a non-trivial chance of complete disaster, and that quite drastic early steps to prevent it would have been justified.

Suggested reading

Could you predict Brexit?

By Tom Chivers

And a Chinese invasion of Taiwan has the potential to be really bad. The superforecasters put together some conditional forecasts as well – that is, predictions of the form “How likely is event X if event Y happens?” So, for instance, if there is a conflict between China and Taiwan, how likely is the US to come to Taiwan’s defence, and how likely would China be to preemptively attack US forces?

The median estimate for how likely the US is to come to Taiwan’s aid if there were an invasion is 83%. So we are talking about a very high probability that a Chinese attack on Taiwan would lead to armed conflict between the world’s two superpowers. They also think it’s about 75% likely that the US would try to sink Chinese invasion ships, and say it’s reasonably likely that China would preemptively attack the US forces in the region if they did attack.

What might the knock-on effects be, if the world’s largest economies end up in a shooting war? Well: the US imports about $470 billion’s worth of goods from China a year. The superforecasters’ median estimate is that that would drop by 20%, or, roughly speaking, $100 billion. That’s the equivalent of the entire economy of Ecuador or Kenya. It would mean a huge blow to the world economy and probably push millions of people back into poverty. Huge US firms such as Nike or Apple would most probably stop manufacturing goods in China, again undoing decades of economic growth that has driven the rise of the Chinese middle class.

And what’s more, it’s very far from obvious that the US would win. If a war were to break out over Taiwan before 2026, the median estimate is that there’s a 57% chance of Chinese victory; if the war were to break out between 2031 and 2035, when China has had another decade to build up its military relative to the US, the estimate is 66%. “Once committed, China would need to prevail in order to save face and not have their government topple,” writes one. “I don’t believe China will start a hopeless war, so the likelihood of winning is over 50% if they start it, says another. 

Suggested reading

China's Communist Party owes it all to colonialism

By Tom Chivers

I asked the forecaster I spoke to if they’d talked about how likely it would be that a US-China conflict might escalate to a full-scale war or even nuclear war, and they hadn’t; they may readdress the question in future. 

The Korean War started because great powers misjudged other powers’ willingness to defend allies and honour their treaty obligations: China and North Korea didn’t think the US would defend South Korea; the US didn’t think China would enter the war directly if the 8th Army crossed the 38th Parallel. It could easily happen again.

And the experience of that war shows that even a limited conflict between two great powers can be terrible – millions dead, years wasted, a country devastated for decades. Taiwan probably won’t spark a US-China war in the next five years, but it might, and if it does it would be a disaster.