What happens if Russia uses a nuclear weapon?
Would a NATO country enter Ukrainian airspace? Attack Russia’s Black Sea fleet? Respond with nuclear strikes of its own?
Forecasted military consequences within 90 days of Russia using a nuclear weapon in the next six months:
- A NATO country offensively entering Ukraine/Ukrainian airspace: 58%
- NATO countries sinking more than 40% of Russia’s Black Sea Fleet: 35%
- An American nuclear strike on Russia: 11%
We also considered 11 other possible effects, including how a hostile detonation might affect American and Russian GDP, and if it would affect the likelihood of war across the Taiwan strait.
If there’s a topic you would like Swift Centre forecasters analyse, or if you’d be interested in our team running a workshop at your organisation, contact us at email@example.com.
Consequences of a nuclear detonation in Europe in the next six months
For 90 days after a detonation (unless otherwise stated)
Earlier this month, we published our forecasts on Russia using a nuclear weapon and how it changes under different conditions. Our forecasters assign a greater chance of a nuclear weapon being used in combat than historical precedent would suggest but, if a nuclear weapon is used, what happens next?
We asked our forecasters — who have strong track records in forecasting geopolitical events — for important events that they believe may take place in the wake of a nuclear strike. After reviewing and refining their suggestions, we gathered the group’s forecasts on 14 potential consequences.
Since our last post, Ukrainian forces have surrounded Kherson, an important port city occupied by Russia, while Russia has pummelled Ukrainian energy infrastructure. Vladimir Putin also claimed that Ukraine was preparing to use a “dirty bomb” – a low-yield radioactive device designed to sow panic – on its own territory, and that it would blame the attack on Russia. Kyiv accused the Russians of concocting the allegation in order to provide themselves a pretext for escalation.
Additionally, Russian state media showed Vladimir Putin overseeing the annual exercises of Russia’s strategic nuclear forces. His defence minister, Sergei Shoigu, stated the drill was preparation for making a retaliatory strike, while Putin said there was “no point” in Russia using nuclear weapons in Ukraine.
How likely is a nuclear attack in the next 6 months?
Our forecasters were asked (a second time) how likely it is that a nuclear weapon be detonated in Europe as an act of hostility before 30 April 2023. (The detonation of a dirty bomb does not count as the detonation of a nuclear weapon)
The aggregate forecast was 4%. That is less than half the likelihood we assigned to the same prospect three weeks ago, as Russia has displayed its commitment to fighting with conventional weaponry. The figure nevertheless suggests a dangerous situation — one that, if it persisted over decades, would make the use of nuclear weapons overwhelmingly likely.
Our forecasters did not generally think that the use of a nuclear weapon would help Russia and Putin meet their objectives. Between them, they cited the prospect of US retaliation, the risk of being cut off by allies such as China, and the increased likelihood of Putin being assassinated or toppled.
Explaining their reasoning for their low assigned probability, a forecaster mentioned that Putin has “mobilised hundreds of thousands of new troops and launched a huge wave of strikes aimed at crippling Ukrainian energy infrastructure. Does this look like the actions of a man who thinks he's out of conventional options?”
One of those alternative options includes blowing up the Kakhovka dam. Such an attack, Volodymyr Zelenskyy has warned, would devastate southern Ukraine’s water supply and leave the nuclear plant at Zaporizhzhia without cooling water. It would be “a devastating step short of a nuclear bomb that could placate hardliners in Russia.”
The same forecaster expressed concerns over Russia emptying Kherson of civilians and Russia-loyal provisional government officials. Kherson could be used as a location for a detonation, the forecaster warned. “This is a remote chance but needs to be continuously monitored.”
Another factor in favour of lower probabilities, as pointed out by several forecasters, was that Putin’s order to use a nuclear weapon might not be followed. On the other hand, another forecaster maintains that being under pressure could place him at risk of making poor strategic decisions.
Likely military responses
Forecasters considered NATO’s potential military response. These responses included entering Ukraine a combative capacity; attacking Russia’s Black Sea fleet; and attacking or blockading Russian assets and territory beyond its immediate borders.
A NATO member country offensively entering Ukraine/Ukrainian airspace (90-day risk)
Of the scenarios presented to them, the forecasters considered the most likely to be NATO member forces entering Ukraine, giving a 58% likelihood of this happening within 90 days of detonation. 90-day risk levels forecasts were given so that they apply between now and August 2023.
However, there was a wide range of responses. One forecaster put the probability at 90%, and another at 1%. Forecasters referred to suggestions from NATO, EU and American officials that a nuclear detonation would incur military action — though they recognised that such comments could have been intended as a deterrent rather than signalling true intent. Forecasters envisaged action on the part of Ukraine’s closest allies in NATO — Poland, the US and the UK — as more likely than action on the part of NATO as a unified whole. Hungary, Turkey and Italy were seen as potential obstacles to a joint NATO mission.
A potential “horizontal escalation”, one forecaster said, would be the West attacking only targets directly related to the detonation: for example, the ship that launched the nuclear missile. “Whether or not [the West stays] out of Ukraine within 90 days of the initial detonation would depend on what Russia does in response to this horizontal escalation.”
The US “would at a minimum provide air support at least in Western Ukraine up to Kyiv”, another forecaster thinks, “even if the US did not fly bombers and fighter jets close to the border with Russia”. Without a nuclear detonation, the probability of a Western incursion over any 90-day period (between now and August 2023) was estimated to be 0.6% by the group.
US strikes against Russian military assets outside Russia (90-day risk)
Of the scenarios they considered, the forecasters deemed American strikes against Russian military assets outside Russia (e.g. their military base in Tartus) to be the second most likely.
The aggregate forecast came to 51%, but there was a wide spread of estimates — ranging from 90% to 10%. The forecaster who put this probability at 10% thought this scenario was “too escalatory” to be particularly plausible, but others felt that the United States would, at this point, have to back up its rhetoric with action. Attacking the base in Tartus, which is the second-largest port city in Syria “risks destabilising an already fragile region with the possibility of the rise of more extreme elements,” one forecaster felt. Another commented that an American strike could encompass an attack on the ship that launched the nuclear missile. The forecaster saw such a response as being in keeping with American dress rehearsals for similar scenarios. “We know from reports of US wargaming exercises involving nuclear scenarios that all participants, from the hawks to the doves, responded in some significant way against Russian military assets, and it would be seen as less escalatory to attack Russian assets outside Russia than inside Russia.”
Kaliningrad being blockaded (90-day risk)
The third-likeliest military response, the forecasters judged, was a blockade of Russia’s Baltic outpost of Kaliningrad. The enclave is geographically separate from Russia proper, bordering Poland and Lithuania. It is home to Russia’s only ice-free European port and is reported to house nuclear-capable missiles.
The aggregate forecast of the probability of the blockade was 44%. Should there be no detonation, the estimated likelihood fell to almost zero. Some thought that it would only come after “Russia escalated after an initial Western response”, while others felt it was an “obvious, relatively non-escalatory step NATO could take”.
Black Sea navigation: 90-day risk of the Montreux Convention being scrapped
A significant obstacle for NATO naval action is the Montreux Convention, the international agreement that governs passage between the Mediterranean Sea and the Black Sea. Under the agreement, Turkey has the power to close the Dardanelles and the Bosporus Straits, through which all ships travelling between the two bodies of water must pass. Any countries sending warships through the Straits must notify Turkey beforehand, with Turkey notifying each signatory to the convention. The signatories include Russia and the UK. On February 27, Turkey announced it was closing the Straits.
They put the probability at 35% – thirty times more likely than if there were no nuclear detonation. The forecasters highlighted regional complexities, with Turkey at odds with its fellow NATO members regarding its conflict with Greece over Cyprus, as well as its ties to Russia. “If Russia detonates a nuke, Turkey may be compliant and permit the US to send more ships into the Black Sea, but unless the Russian Black Sea Fleet has already been sunk and their Air Force defanged, it is at huge risk,” a forecaster wrote.
NATO member countries sinking >40% of Russia's Black Sea Fleet (90-day risk)
The scrapping or ignoring of the Convention may occur in the run-up to an attack on Russia’s fleet in the Black Sea. On aggregate, the forecasters put the probability of NATO members sinking more than 40% of Russia’s Black Sea Fleet at 35%. (With no nuclear detonation, the probability was put at close to zero). The attack was broadly seen as too escalatory to be particularly likely, but forecasters referred to the option having been discussed in policy circles. “Some, including David Petraeus” – the former director of the CIA – “have said that the Black Sea Fleet will be sunk if Russia detonates any type of nuclear bomb as a hostile act against Ukraine,” said one, commenting that it would be “wise to be decisive yet leave enough material and troops that Russia would not want to endure further losses. The NATO action would also need to take into account how best to create dissent against Putin within Russia.”
The West has shown with its existing economic sanctions against Russia that it is willing to consider significant non-military responses. One forecaster suggested that a remaining diplomatic lever left is to label Russia a state sponsor of terrorism. This would make it easier for victims of Russian violence to sue Russia in American courts, potentially incurring vast sums in damages. Among other effects, the labelling would also make it harder for third-party countries to trade with Russia.
Russia being designated a state sponsor of terror by the US (90-day risk)
The aggregate prediction was 40%, with the likelihood much lower (4%) should there be no nuclear detonation. Forecasters wrote that the Biden administration does not seem interested in this step, perhaps saving it as a chip to use in peace talks. “Russia is de facto already facing as great or greater restrictions,” wrote one forecaster, “than the countries currently listed as state sponsors of terror (Cuba, Syria, North Korea, Iran). The designation could be immediate but I'm not sure if it would even be a priority compared to the kinetic responses that would occur.”
A cyberattack shutting off most of Russia's electric power (90-day risk)
Another Western response could be to launch a cyber-attack on Russian infrastructure. Forecasters thought this step to be relatively unlikely (13% in the event of a detonation, less than 1% otherwise). They commented on the technical difficulty of knocking out the majority of Russia’s power; the likelihood that this option would be seen as escalatory rather than horizontal; the possibility that the toll taken on freezing civilians would make such an attack a war crime; and the Biden administration's probable unwillingness to open up American civilians to a response in kind.
Nuclear retaliation and nuclear war
Would NATO countries respond in kind to a nuclear detonation? For decades, the prevailing approach has been to maintain strategic ambiguity — but the use of a nuclear weapon would force the West to show its hand.
States other than Russia using a non-test nuclear weapon by 2033
On aggregate, the forecasters put the probability of this eventuality at 13% (and 1% with no detonation in the first instance). This is much higher than the usual base rate of the use of a non-test weapon. With NATO war-gaming not having produced a clear answer, forecasters thought that the Biden administration, fearing nuclear war, would be strongly inclined not to respond in kind. But they did not rule out other countries using nuclear weapons in separate conflicts.
American nuclear strike on Russia (90-day risk)
Again, this act was seen as too risky and aggressive to be particularly likely: 11% if it followed an initial detonation, close to zero otherwise. But that 11%, forecasters pointed out, was much higher than the base rate for nuclear weapon use. They noted that an attack on a military target, including one that gave Russia plenty of warning, was more likely than an attack on a civilian population.
The major risk of a tit-for-tat strike, of course, would be the development of the conflict into full-blown nuclear war, which the forecasters went onto analyse next.
Nuclear war between America and Russia that results in at least 1 million fatalities (90-day risk)
It is worth recalling that this troubling scenario, like the others considered by our forecasters, rests on there being a first instance of a nuclear weapon being used – a prospect they gave a likelihood of 4.3% over the next 6 months. In that event, the chance of a large-scale nuclear war – one that, as per the question above, would kill at least a million people – was forecast at 5% probability. In the event of no initial nuclear weapon use, they give less than 0.1% of this starting over any 90-day period (between now and Augst 2023).
Forecasts ranged from 10% to 0.25%. A wise strategy on NATO’s part, wrote a forecaster, “would be to leave enough buildings standing, as well as military capacity, that Russia would still have something to lose by continuing the escalatory cycle. I also think an initial US response would prioritise the destruction of fighting capacity around the Ukraine and Black Sea regions, and try to minimise loss of life. Putin may be toppled from within, and even if he is still in power, there would be tremendous pressure to avoid Armageddon.”
Our forecasters also considered what the economic impact of a nuclear detonation would be, and provided their expectations for Russian and US GDP under this scenario.
US GDP in 2023
Russian GDP in 2023
In line with Goldman Sachs projections, our forecasters put American GDP in 2023 at $23.5trn. In the event of a nuclear detonation, the aggregate forecast was $21.3trn – a drop of about 10%. One forecaster explained that their estimate took into account both a 10% chance of a full-blown nuclear war, in which case GDP would “probably be zero”, and a more likely reduction of about 4%, “to account for all the chaos that would still happen as a result of a tactical nuke: globalisation getting more unstable as a result of Russia being cut out of the world economy, supply chain disasters, vastly more expensive gas and oil, etc.”
The forecasters envisaged a more severe contraction befalling Russia: $1.35trn GDP in the event of a detonation, versus $1.63trn in the event of no detonation. “If Russia does detonate a nuke, I expect their economy to be devastated,” wrote one. “The immediate retaliation is likely to severely damage what's left of their military. They will be a pariah state, but still find trading partners in China, Africa and the central Asian republics.” Russia would also face the risk of some states breaking away, wrote the forecaster.
Vladimir Putin leaving office (90-day risk)
The gamble of using a nuclear weapon would probably affect Putin’s hold on power. As the forecasters pointed out, Putin’s deployment of a nuclear weapon might suggest he is losing badly enough in Ukraine to make him vulnerable at home. With no detonation, they put the chance of Putin leaving office within a 90-day period at 4.2%; with a detonation, 44%. This figure hides a broad range of responses, with one estimate at 79% and another at 10%. Putin was not seen as in immediate danger, but forecasters thought the use of a nuclear weapon would raise his risk of being ousted. “Even normal people without significant amounts of power might even try to overthrow Putin if he uses a nuke,” one forecaster commented.
A war across the Taiwan strait by 2027
It has been theorised that a weak Western response to the invasion of Ukraine might invite an emboldened China to make good on its threats to Taiwan. What would really happen?
Although they expressed significant uncertainty, the forecasters put the chance of a war across the Taiwan Strait in the next 5 years at 26% with no nuclear weapon detonated in Europe, and 21% in the event of a detonation. “Both the US and China may be spooked by the prospect of where war could lead humanity,” wrote one forecaster, but another said Xi’s actions would be informed by how the West responded to a Russian escalation. Overall, though, the forecasters shied away from making particularly confident predictions.
There is a wide array of plausible Western responses to a Russian nuclear weapon. Our forecasters see a non-nuclear military response as significantly likely, and a nuclear response as much less probable – though nonetheless still serious.
If there are topics you'd like the Swift Centre's panel of forecasters to tackle, please let us know: firstname.lastname@example.org