Towards a New Nuclear Arms Race? Putin, the Breakdown of Nuclear Treaty Limits & MIRVs
With the invention of the atomic bomb, humanity came up with a weapon capable of posing an existential threat to entire nations or civilizations. Then through the decades-long Cold War decided to go about building tens of thousands of them. While the nuclear powers also looked for ways to make their growing arsenals even more destructive. Yields increased, warheads shrank, and relatively quickly the technology was developed to enable even individual missiles to carry multiple nuclear warheads. It was a nuclear arms race that continued apace until a series of international agreements first slowed and then started to reverse the apocalyptic build up. Over the course of a few short decades the US, Soviet and later Russian nuclear arsenals were reduced to a tiny fraction of their former size.
Some systems perceived as being particularly destabilising were banned outright. And the world as a whole probably let out a sigh of relief as warhead counts trended downwards. Now however, things are changing. Russia has suspended verification inspections under the most important US/Russia arms limitation treaty, New START, in 2022. And the agreement as a whole is now just over a year from expiring entirely. We've started to see countries, including Russia, roll out new generations of more capable nuclear delivery systems. Globally warhead counts are once again on the rise, and more nations appear to be moving towards fielding potentially destabilising technologies like Multiple Independent Re-entry Vehicles.
And so today we are going to focus on one aspect of this new potential arms race. With the potential collapse in the restrictions on large strategic missiles and the warheads and MIRVs that make them so dangerous. To do that we'll start by looking at the technology of MIRVs themselves, their emergence during the Cold War, and what it is about their effect on the calculus (if you like) of a potential nuclear war that led to them being de facto restricted in the first place.
We'll then look at that nuclear treaty regime, what it achieved, and what it might mean that Russia has suspended participation in that treaty for more than a year at this point. And that the agreement as a whole is now just over a year from expiring entirely. Finally, we'll close out by looking at how the nuclear powers, from Russia and the United States to India, China, the UK and France, appear to be preparing for the coming years of nuclear competition. But before we jump into that particularly happy topic, allow me to welcome a new sponsor. Odoo is an all-in-one family of management software covering a range of purposes. With 45+ business applications, there's kind of an app for every need.
And their website builder is a pretty nifty tool if you want to build a website quickly easily and efficiently - even if you've never done it before, like me. Now I've long thought that Perun might benefit from having a website, and thanks to Odoo, now it's getting one. I started by choosing a business type: defence economics consulting wasn't a default option, but I was able to get fairly close. And then went about choosing an objective. I had the option to feed in my logo to help pre-generate some designs and colour schemes and went from there.
I selected the pages and features I needed and finally chose one of the themes. On the website you can add, drag and drop blocks, text and images to personalise it. And Odoo makes it easy to add visual content, pulling images from your PC or a built-in royalty-free library. You can customise those images and animations and refine everything from your typography to layout. Where it's usually as simple as using a layout grid to cleanly organise all your elements.
And perhaps the best part of all, your first app of Odoo software, whether it be the website builder or another application, is free for life with unlimited hosting and support. And they offer a free personalised domain for one year as well. I hope you are looking forward to seeing the results when the site goes live. But for the moment if you are interested, please do check out todays sponsor Odoo, you'll find the link in the description. And so with that said, let's get back to it.
The early stages of the Cold War saw a very rapid build-up in atomic arsenals. There was a brief window during the US nuclear monopoly where serious discussions were given to potentially banning nuclear weapons entirely. But after the Soviets tested a device of their own in 1949, it was basically off to the races. By the early 1960s the US Single Integrated Operational Plan 1962, basically the nuclear war plan, called for the deployment of nuclear weapons equivalent to about 7.8 gigatons worth of TNT against targets across the communist world,
with estimated casualty counts in the hundreds of millions and collateral damage on a national scale. The tiny nation of Albania for example was to be nuked repeatedly because it was home to Soviet air defence radar. The Soviets meanwhile operated on the assumption that World War Three would probably mean the Red Army fighting its way through a nuclear hellscape.
This was an era where nuclear warheads were being put on just about everything. Nukes on air-to-air missiles, nukes on artillery shells, there was even an American concept for a nuclear powered weapon that would carry and deploy nuclear weapons. Thankfully the concept of nuclear powered nukes was recognised as being too bat-shit crazy, even for the Cold War era. At least that is, until the concept was later revived by Vladimir Putin's Russia.
According to estimates by the Federation of American scientists, the US nuclear arsenal would peak at 31,255 warheads in 1967. The Soviets would peak in 1986 at 40,159. In total, global inventories would crest above 70,000. And a lot of those apocalyptic warhead counts were going into the kind of systems we are concerned with today, long-range ballistic missiles equipped with multiple independent re-entry vehicles. Initially the nuclear force focus, especially in the United States, was with the bomber fleets. In the 1950s and 60s the US built up a massive bomber fleet under the auspices of Strategic Air Command, with one of those, the venerable BUFF, still being in service today and still nuclear compatible.
The problem for the bombers though were twofold, although they both came from the same line of technological development. The first was that surface-to-air missiles and air defences in general were just getting too good. Even the prototype American XB-70 Valkyrie, capable of exceeding three times the speed of sound and quite possibly one of the most beautiful aircraft to have ever flown, would have had questionable survivability against ever-improving Soviet air defences.
And on the other hand, while missiles could potentially be used to shoot down nuclear bombers, they could also be used for nuclear delivery themselves. The world got its first ICBMs in the 1950s with the Soviet R-7 and the American Atlas. By modern standards these weren't particularly good weapons and the Atlas was retired by 1965, but the concept was proven and Pandora's box was open. From there, budgets were extensive and technological development rapid. Everyone rolled out a sequence of designs that improved on just about everything, from fuelling time and readiness, to reliability, to reduced cost, to increased accuracy, and even the ability eventually to move long-range strategic missiles carrying nuclear warheads into the maritime domain, with the birth of things like the Submarine-Launched Ballistic Missile.
And with it, another leg in what we now know as the "nuclear triad". But arguably two of the most significant innovations in determining just how stuffed the world would be if a nuclear exchange ever occurred didn't have to do with the missiles themselves, so much as the warheads they carried. The first was obviously the development of the hydrogen bomb and later its miniaturised equivalents. This enabled nuclear powers to do a range of things, some of which were horrifying but arguably practical, and in other cases just horrifyingly stupid. These Soviets for example would famously build the Tsar Bomba, a thermonuclear device rated for up to 100 megaton yield, which they detonated at about half that value.
Proving presumably to the United States and its allies that Soviet industry was capable of matching or exceeding them in a variety of areas. Including the ability to spend enormous amounts of money developing ego-extension super weapons with deeply questionable practical military value. But on the more practical side, the ability to squeeze ever more atomic boom into smaller and smaller packages made lifting more than one warhead per missile practical.
That opened the door to the concept of the Multiple Independent Re-entry Vehicle, or MIRV, where one missile carried multiple warheads, each capable of going after different targets. The US was first to the punch, coming up with a new version of the Minuteman missile that could mount 3 warheads instead of the previous one. The Soviets already more or less had a heavyweight ICBM ready to go, and so MIRVed up their R-36.
It had originally been designed to carry a serious multi-megaton warhead, and so the design had plenty of throw weight available to switch over to some still pretty chunky MIRVs. Submarine-launched ballistic missiles also got the MIRV treatment. Polaris would eventually carry 3 warheads for example. And within a few short decades the numbers climbed and climbed.
By the 1980s the Soviets had re-configured the latest versions of R-36, now NATO designation SS-18 Satan, capable of carrying a casual 10 MIRVs in the mid to high hundreds of kilotons range, which was in the process of being matched by the then new American LGM-118 Peacekeeper, mounting ten 300 kiloton MIRVs. While NATO's de facto front-line SLBM, Trident, was rated for up to 8 MIRVs. To put all this into perspective, if we take a 750 kiloton per warhead estimate for the Soviet SS-18, by the end of the Cold War it was possible for a single heavyweight Soviet missile to deliver something like 50 times the yield of the device used on the city of Hiroshima against each of 10 separate targets. OK, so if that's the underlying technology behind MIRVs, why are they so significant? And what does that potentially mean for various nuclear scenarios? When we are looking at a scenario involving nuclear armed nations and the potential risk involved, there are a couple of elements we might want to look at. We might care for example about how many warheads are involved and their potential yield, because that defines how damaging a potential exchange could be.
Would it merely result in human tragedy on a massive scale? Or does it reach a point where civilization starts swapping out the dollars and euros for bullets and bottle caps? But if what we care about specifically is risk and strategic stability, that is the chance that you actually end up getting a nuclear exchange, including by accident, then we probably care more about factors like the relative balance between the two sides. Whether the weapons being used have a first-shooter advantage so to speak. And what broad launch policy the countries in the scenario are likely to adopt.
Something which, again, is likely to be shaped by the nature of the weapons being used. At one extreme for example, you might have a pre-emption policy. You fire your nukes if you get intelligence that suggests your opponent is about to fire theirs, even if they haven't yet. That might give you the first-shooter advantage in the exchange that follows, but also risks you destroying the world because of faulty intelligence. A step down then would be launch on warning.
You launch your missiles as soon as your sensors detect that the opponent has launched theirs. That increases the odds you can launch your missiles before they are destroyed by whatever is incoming, but history absolutely includes examples of nuclear warning systems giving false alarms. So a particularly twitchy launch on warning policy might turn a minor technical glitch into one of the most costly support tickets of all time.
Finally, at the conservative end of the spectrum, you might have a policy of retaliation after ride out. You only launch your nukes after you are very sure that you've been nuked. And you do that by waiting for the fireballs, and then letting rip with whatever you have remaining. From the point of view of trying to minimise the odds of a nuclear catastrophe, we'd probably prefer that as many countries as possible adopted a very conservative launch policy. But the basic capabilities and performance characteristics of certain nuclear weapon systems can push them one way or the other. And as with almost any situation where we are talking about strategic stability I think it pays to have a scenario, where old rivals Kiwiland and Emutopia are trying to understand the impact of one of the most critical factors in how a potential nuclear arms race between the two of them is likely to play out.
And that critical factor is maths. The core issue here is that while nuking something with a ballistic missile might sound like a pretty definitive way to make sure you destroy it (and in a lot of circumstances it is) missiles aren't perfect. There are technical risks with the missile itself, it could fail to launch, break up in flight or the warhead could fail to detonate. It might be shot down by an opposing missile defence system, or not land close enough to a hardened target to destroy it even with the help of a nuclear blast.
Or alternatively, the target might be fleeting or unknown. If you are targeting an opposing road-mobile ICBM that you know has four firing points, you might not want to settle for guessing exactly which one it's in. You slam all these factors together in a model and you can get your single-shot Probability of Kill. And of course because these issues can be multiplicative, even if the individual numbers aren't that large, the probability can fall off relatively quickly. Take a 10% chance of technical failure, 10% chance of interception, and a 30% chance that a warhead doesn't land close enough to a hardened target to actually destroy it, then the probability of destroying that target with a single missile in this very simple model is about 57%. That might seem like a relatively low success chance for a literal strategic nuclear weapon, but if you think on the modern battlefield of how many tens or hundreds of thousands of rifle rounds have to be fired per infantryman hit on average, or how many drones have to be used on average per vehicle destroyed, a 57% success rate when you are talking about destruction on this scale is actually pretty scary.
At the same time though, if your game of choice is nuclear war historically planners haven't really been happy with numbers like 57%. If Kiwiland wants to be able to knock out Emutopia's missiles on the ground before they can launch for example, they probably want to be sure they get all of them. Since this is a scenario where bad RNG might potentially result in cities disappearing. During the Cold War era this often led to US nuclear war planners demanding very high confidence levels for the destruction of certain Soviet targets. And that in turn often demanded some pretty silly warhead counts.
When an investigation was done into America's nuclear war plan, the Single Integrated Operational Plan, for the new George H.W. Bush administration, they found a couple of examples that showed just how determined the targeting staff were for the Soviet Union not to be allowed to have an air force anymore. "One of the first, which they'd noticed ahead of time in the SIOP's National Strategic Target List, was a Soviet bomber base in Tiksi, inside the Arctic Circle. It wasn't even a primary base, it was a dispersal base, where the Soviet planes would land after dropping their bombs on American targets, and the climate was so forbiddingly cold and windy that the base couldn't be used for more than half the year. And yet it turned out the SIOP called for firing 17 nuclear weapons at a 5 mile radius around the base - regardless of the season - including 3 Minuteman II ICBMs, each carrying a 1.2 megaton warhead." Another example was an airport just outside Moscow protected by the city's anti-missile defences.
Against that target, to secure a high enough probability of kill the plan assigned almost 70 warheads. Some of that Cold War-era targeting process was clearly guided by a chilling kind of military logic. While some of it probably didn't make sense even then, if someone had actually stepped back and asked whether it really made sense to spend billions of dollars in 2020s terms targeting a single airport in Moscow. But if you hold on to just that initial premise, that firing a single nuclear warhead against an opposing harden target is inevitably going to give you a PK value of less than one, you'll immediately understand how you might end up with comparatively strategically stable situations even when you are dealing with a stand-off between two nuclear states. For example let's imagine that Kiwiland and Emutopia are facing off with one another using comparatively bargain basement nuclear arsenals. They each have 10 silo-based ICBMs, each carrying a single warhead.
In that scenario, even if Emutopia can find a moment where the Kiwis aren't on alert and there's no risk of the missiles leaving their silos before the incoming missiles arrive, even a full send counter-force nuclear sucker punch probably wouldn't be expected to destroy all of the target silos. This, as you can probably imagine, is a deeply suboptimal strategic situation for Emutopia. Their missiles are gone, some of the Kiwis' aren't, and the next step is probably a very angry phone call that begins with the diplomatic equivalent of "WTF mate." And the Kiwis going on to note that Emutopian cities are lovely this time of year, and it be a real shame if something happened to them. In other words, in this kind of simplified evenly-matched scenario shooting first is kind of for idiots, because with every shot you fire you're actually increasing the disparity of remaining firepower in your opponent's favour.
Not to mention all the other factors that might make that kind of strategy absolutely idiotic, like the rest of the world responding to your nuclear first use. In this kind of scenario, civilian populations still have to live under the spectre of being destroyed in nuclear fireballs. But they can live with the small consolation that it doesn't make sense for either side to be the one to pull the trigger first on a nuclear exchange. It also means that for either side to have decent odds of pulling off a nuclear first strike against their opponent's nukes, they need to build many more missiles than their opponent has targets which will probably be both expensive and noticeable.
Making it even more likely the two sides will remain locked in a horrifying but stable mutually assured destruction counter-value face-off, as opposed to nuclear war planners getting bright ideas about potential counter-force options. Great, so far so good, the maths checks out in favour of strategic stability. And then someone has to go and invent the MIRV and mess it all up. Because now a single missile being launched doesn't mean a single warhead being launched, it might mean 3 on the Minuteman III, 10 or 12 on the old Peacekeeper, or somewhere between 10 and 16 on some of the Russian heavies. And suddenly all the maths changes. Missile defence also becomes a lot harder and much less economically attractive.
In a single warhead missile scenario, you might have an opponent using one missile to fire a warhead that needed two or three interceptors engaging it to have a good reliable chance of interception. Given how expensive ICBMs and IRBMs can be, this isn't a particularly catastrophic situation for the defence. But if each of those expensive incoming missiles is carrying 10 warheads plus decoys, things become much harder for the missile defence guys. If you can't destroy the missile before the MIRVs separate, you might now be talking about as many as 30 interceptors for every one incoming missile. And in all but the most extreme cases of economic disparity, say the United States versus North Korea for example, no one in that kind of scenario is going to be able to deal with the enormous costs required to build up the missile defence shield to counter the opponent's sword.
This is why I always get very confused every time Russia complains about NATO deploying a missile defence system and describes it as "destabilising" or "potentially provocative". The Russian nuclear arsenal is extensive and their missiles are among the most heavily MIRVed in the world. There is no missile defence system, either existing or conceptually possible in the near to medium term, that could possibly deal with a Russian nuclear attack. And what is there, is more scoped for the kind of threats you might see from the likes of Iran or North Korea. If you are doing investment planning for a nuclear force then, there's probably two main ways you could potentially use this MIRV superpower. The ability to more economically deliver a lot more nuclear warheads than you could previously.
Firstly, you could use that technology to fulfil that underlying nuclear deterrence mission, if not on the cheap, certainly the cheaper. If your main goal is just to hold opposing counter-value targets at risk, having MIRVed missiles allows you to achieve the same amount of threat with considerably fewer missiles. This might be particularly important if you are a country like France or the United Kingdom with all of your city busting capacity focused just on the submarine arm. Because now you can lift the number of warheads you have out there in the ocean, presumably survivable, acting as your deterrent at any given time, without either building a massive fleet of submarines, or a fleet of massive submarines. So used in this way, you could argue MIRVs are a massive gift to navies and treasuries around the world, because they are arguably a significant part of what makes a survivable continuous at-sea deterrent possible. The place where MIRVs potentially become a bit more chaotic when it comes to nuclear war planning is when you start talking about using them in the counter-force role.
Remember here the basic goal is for me to nuke your nukes with my nukes before you nuke me. And when you start putting 3, 10, 16 warheads on missiles that maths we discussed before fundamentally changes. You can go from a scenario where on average one missile launched is going to destroy less than one opposing silo, to one where a single missile is expected to destroy multiple silos.
Plus if your opponent is also using MIRVs, the amount of fire power you take off the table whenever you destroy a missile in its silo also increases. Against an un-MIRVed ICBM force, destroying one silo or one road-mobile launcher takes one warhead off the table. With MIRVs however, that same hit could be worth 3, 9, 12 or 16 enemy warheads. Now potentially you might be in a positive exchange scenario, where you expect to use fewer missiles and warheads than you end up destroying. Now suddenly you're in a world where Han Solo might have a bloody point. And while this might sound like a purely theoretical discussion, just an on paper vulnerability, for historical nuclear planners it was very much a real concern.
The following quote for example comes directly from a US Congressional Budget Office document dated June 1978, titled Planning US Strategic Nuclear Forces for the 1980s. "The most immediate problem is the growing ability of increasingly accurate Soviet multiple-warhead ICBMs to destroy US silo-based Minutemen. By the middle of the 1980s, Soviet missiles might be accurate enough - that is accurate to within about 600 feet of their targets - to destroy more than 90% of the US land-based missile force. The predicted vulnerability of the Minuteman missile force has led to questions about both the future adequacy of the US assured destruction capability and the desirability of maintaining a strategic triad that includes land-based missiles. The growing Soviet ability to attack US ICBMs has also raised the question about whether or not the United States should acquire a matching counter-force capability." That document also raised, despite the enormous number of warheads in the inventories of both sides, the prospect of a potential Soviet first strike.
This was actually the plot line of the classic 1979 film First Strike, which I'm sure all of you have seen. But just in case - spoiler alert here for a 45-year-old film - which is basically 50% dramatisation of a Soviet counter-force first strike against the United States nuclear deterrent (notably produced with the full cooperation of the United States Department of Defence and RAND Corporation) which results in about 80% of US strategic forces being destroyed in that first strike, prompting the President to open negotiations with the Soviets. The second half of the film is basically a series of interviews with experts explaining to the American public why billions of dollars should probably be spent on the MX missile system. I'm not sure how many viewers who expected they were getting a nuclear war movie would have responded when it suddenly morphed into a series of budget justification arguments. But the core point here is that while that might sound like quite a fantastical scenario, it wasn't just finding root in fiction, you can see fears of something like that surfacing even in some US documents from the time.
To quote from that previous document, "In some future crisis, Soviet leaders might decide to use a fraction of their large multiple-warhead ICBMs to launch an attack that could destroy the US Minuteman force, but would avoid the destruction of US cities. Perhaps 5 million Americans would be killed by such a counter-force attack, but most Americans would remain alive as long as the Soviets refrain from launching direct attacks on US cities. And thus, as long as US cities remained intact, American leaders would have a powerful incentive to avoid direct attacks on Soviet cities.
If attacks on Soviet cities were the only response available to the President, the US retaliatory threat might not be credible to Soviet leaders who might gamble that the US could be coerced by an attack on the Minuteman force. Thus to enhance the credibility of its nuclear deterrent, The United States might want forces capable of destroying Soviet ICBM silos in a second strike. An ability to destroy Soviet ICBMs held in reserve after an attack on US forces could also prevent the Soviet Union from gaining a relative advantage in strategic capabilities surviving a missile exchange." The bottom line here is that when you combine MIRV technology (particularly accurate MIRV technology) with things like silo-based ICBMs, all else being equal you are shifting the balance in favour of the nuclear offence over the nuclear defence. And increase the potential incentives for one side or another to potentially want to either shoot first in order to catch their opponent's MIRV missiles on the ground, or to adopt more hair-trigger style launch policies to make sure their missiles survive an opposing counter-force attack.
That makes achieving safe strategic stability in a scenario with lots and lots of MIRVs more difficult than in a world where only single-warhead missiles exist. With one main counter-measure being finding systems that can provide deterrence and are survivable even if an opponent launches a first strike using MIRVed weapons. The key example that comes to mind there is the ballistic missile submarine. Though they can potentially come with their own implications for strategic stability. With each submarine potentially representing a lot of nuclear eggs in a single proverbial basket, and increasingly accurate MIRVed SLBMs potentially providing a low-warning counter-force option all of their own. Now perhaps understandably, at some point during this process both the United States and the Soviet Union appear to have looked at themselves and realised they were both sinking budgets equivalent to a small national economy into piling up doomsday weapons with enough combined nuclear firepower to destroy every target of note in the opposing country more significant than the Southport service station.
This would apparently eventually prompt them to ask whether or not this was an optimal course of action, both from a budgetary and survival of humanity perspective, but that wouldn't take place overnight. The basic game theory of treaties is that you are only likely to get one when it's in the interest of all parties to have one. And that alignment would slowly come into focus later in the Cold War, with the testing of nuclear weapons rather than nuclear weapons themselves being one of the first targets for treaty restrictions.
Global nuclear testing ramped up massively in the latter part of the 1950s, with the United Kingdom, United States and Soviet Union setting off 33 devices in 1956, 55 in 1957, and 116 in 1958. In 1958 against a backdrop of other factors and pressures worldwide, the Soviet Union announced that they would be ceasing nuclear testing, as long as everyone else did the same. In Washington and London some might have regarded this as a bit of a cheeky gimmick, given that the Soviets had just completed a major nuclear testing series, while the US was about to kick off the Hardtack series of tests and the UK hadn't yet completed its hydrogen bomb testing. Once those hurdles were cleared however, a moratorium of sorts did go into effect and there were no devices tested in 1959. At least that is until nuclear newcomer France decided to ruin the nice chart with three detonations in 1960 and it was back on like Donkey Kong.
There were 71 detonations in 1961, and 178 in 1962. But from that experience you can at least see the concept of how nuclear arms limitation agreements might evolve. The 1963 Partial Test Ban Treaty (which the French didn't sign) prohibited the testing of nuclear weapons in the atmosphere, outer space or underwater.
If states were going to test nuclear weapons from here on out, they were going to have to bury them first. The Non-Proliferation Treaty was signed in 1968. Yes, China and France wouldn't join it until 1992 and, no, it wouldn't prevent some countries like India and Pakistan later becoming nuclear powers. But it was a case of another internationally agreed restriction on nuclear weapons, with the signatories basically declaring that the nuclear club was now closed and would not be accepting new members. But the treaties and systems we are most concerned with this episode have to do with those long-range strategic nuclear weapons, warhead count, long-range missiles, and of course MIRVs.
Some of the first real restraints on those came as a result of the Strategic Arms Limitation Talks in the early 1970s, aka SALT I. This tried to cap both the growth in US and Soviet offensive systems, but also an arms race in defensive anti-ballistic missile tech. And if you're wondering how could you possibly make a functional nuclear missile defence system using 1960s and 1970s technology, the answer was nuclear weapons. The system you see on the right there for example is the short-range Sprint missile. The missile would sprint towards incoming warheads, going zero to Mach 10 in the space of about 5 seconds, with near is enough is good enough accuracy levels given the W66 thermonuclear warhead on board.
Under the agreement, both the Soviets and United States could put anti-ballistic missile defences around their capital city and one ICBM field. That arguably favoured the Soviets, given they would be protecting the major industrial centre and transportation hub of Moscow while the United States would be protecting Washington DC. And in the end the US didn't actually bother building the ABM installations that it was allowed under the agreement. On the offensive side, SALT I essentially capped the number of long range strategic ballistic missile launchers at existing levels, and it set specific caps for certain equipment categories. There were a lot of things that SALT I didn't do. It didn't roll back the hilariously high number of launchers already in existence.
It didn't regulate warhead count, which was a problem given that MIRVs were now a thing. But it did at least kind of stop the economic bleeding by placing upper limits on at least some systems types. And it had both sides publicly acknowledging the sovereignty of the other and the principle of non-interference.
Neither side ever really wanted nuclear war. And some Soviet theorists were convinced that if they just stayed the course long enough, eventually the Cold War would end as the tide of history caused one side to collapse. In a sense they would ultimately be proven kind of right, even if they got the side wrong. But following on from the success of SALT I, there appeared to be genuine momentum behind the idea of nuclear arms limitation, which gave us SALT II. Signed by Carter and Brezhnev in 1979, SALT II wasn't just about limiting the nuclear arms race, it was about rolling it back. It would have placed a tighter cap on Soviet and American nuclear delivery systems, banned essentially new missile programs, and had a number of specific limitations on things like the longest ranged and MIRVed ballistic missiles.
There was some systems horse trading where both sides were going to be able to keep the assets they valued the most. The US was able to keep nuclear-tipped cruise missiles and the new Trident SLBMs for example, while the Soviets were able to protect their heavy throw-weight ICBMs like R-36. SALT II would never actually end up being ratified.
In part because the Soviets decided to invade Afghanistan, a manoeuvre that has never gone badly for any empire ever. But even though the agreement wasn't ratified, both sides ended up kind of following many of its provisions anyway. And eventually negotiations that start between Reagan and Brezhnev in the 1980s end up with a new agreement being signed in 1991. This was the Strategic Arms Reduction Treaty, or START I, and finally the world got hard reductions in the number of delivery systems and also the number of deployed warheads. Relevant delivery systems were capped at 1,600 and deployed warheads at 6,000 each.
Still enough to be sure to glass every city between Kamchatka and St Petersburg, but at least the numbers were going in the right direction. And START I also created an extensive and expensive system of verification and inspections to make sure that both sides were following those limits. There was a bit of a false start attempt at further reductions with START II in 1993 between Bush and Yeltsin. But Russia only ratified the agreement in 2000, and did so conditionally on the Americans remaining in the ABM treaty, which they pulled out from in 2002, making the whole thing mostly moot, at least from a legal perspective. There was an additional agreement, SORT, reached in 2002, but the next big one came in 2010. The aptly named New START, signed in 2010, was intended to replace START I which expired in December 2009.
It initially gave both sides 7 years to strip their forces back to a new treaty-enforced low. No more than 700 relevant delivery systems, so ICBMs, SLBMs and nuclear equipped heavy bombers. And no more than 1,550 deployed warheads for them. Which if you're doing the maths at home, means that each delivery system could on average only have a bit over two warheads. This would therefore mean not just reducing the number of missiles and delivery systems, but also mounting considerably fewer warheads on them than many of these MIRV-capable systems were capable of carrying.
As one US document put it, stating the relatively obvious, "The Russian Federation has the capacity to deploy many more than 1,550 warheads on its modernised ICBMs and SLBMs as well as heavy bombers, but it is constrained from doing so by New START." As you can see from the chart on the right there, the Russians were actually already under the treaty limits for deployed warheads as early as 2011. They would sail above them as they surged their number of deployed warheads after 2014, but by the time the grace period ended and it had become clear that the US and Europe wouldn't really do that much about the whole Crimea thing, the Russians join the Americans below that dotted line. Extensive verification systems, including mutual inspections, were put in place. And in February 2021, only a year before the beginning of Russia's Special Military Operation, Moscow and Washington would agree to an extension, stretching the sunset of the New START agreement out to February 2026. This cascade of restrictions, reductions, and of course budget cuts, had a pretty significant impact on the maths of US strategic force design.
After all, if you only have a treaty allocation of two or three warheads per missile, there's going to be much less of a premium placed on having missiles that can carry lots and lots of MIRVs. Yes, you might be able to use some of that throw weight to carry more decoys for example, but the value is significantly diminished. And with the US submarine force getting much of the priority on warhead count, one of the biggest casualties of all of these agreements was the US Peacekeeper missile. The Peacekeepers would be decommissioned and destroyed, and the US would roll back to the smaller, cheaper, much older Minuteman design.
Those Minutemen that remained would then be reconfigured to only deploy with a single warhead rather than the previous three. All this was very significant not just from the perspective of what it meant for strategic stability, but also because whenever Russia shows off a heavyweight ICBM with a bunch of MIRVs on it and says "The West has nothing like this," I think it's worth remembering that that statement while technically true isn't because the West is incapable of building these systems. They did build them. And then through a process of negotiations and budget cuts decided to destroy them.
The end result of the New START limits was a nuclear landscape that looked very, very different than it had only a couple of decades earlier. Arsenals were massively reduced, and there was a significant roll back in the number of weapons that were perceived to place a premium on striking first in a counter-force fashion. The accurate ICBM forces in particular now had fewer warheads with which to go after opposing silos. And each opposing silo on average was now in turn going to be worth fewer warheads.
You certainly can't say the world was safe in a universe where thousands of nuclear warheads still existed, but compared to the arsenals, hair triggers and incentives of the 1980s it was almost certainly safer. Now however, arguably for the first time in more than half a century, things might be going in the other direction as the treaty arrangements start to fray and we may be witnessing something of a return of the MIRV. In 2022 Russia announced it would no longer be allowing US inspectors to verify and inspect its nuclear arsenal under the provisions of the New START treaty. They cited factors like US sanctions and unspecified breaches of treaty limitations by the United States, and slammed the door shut on further inspections. Russia did say they would continue to comply with the treaty limitations on delivery systems and deployed warheads, but with the verification system now shut down this was basically a case of "trust me bro" at a geo-strategic level. And given that just a couple of months earlier Russia had been professing that they wouldn't invade Ukraine and that was all just American fake news, trust was perhaps at a bit of a low in the latter part of 2022.
Ultimately international treaties aren't magic. This isn't a video game where you take a massive stability or economic hit every time you break an agreement. And so keeping compliance in place is often a matter of trust, relationships, incentives, and occasionally coercion. And the situation would decline further in 2023 when President Putin announced on state TV that Moscow would be "suspending" its participation in the New START nuclear arms treaty, but not actually withdraw from it. At the time Russian state media also carried comments from a Russian politician stating that Russia could "denounce the treaty" if the United States continue to "Ignore Russian calls to reconsider the way it is implemented."
Putin's announcement wasn't exactly greeted with enthusiasm around the world. A member of the Science and Security Board at the Bulletin of the Atomic Scientists said at the time, "The loss of agreements will increase uncertainty and the chances of misunderstanding, inflate threat perception, and fuel an accelerating arms race." Another commentator, a co-director at the Non-Proliferation Program at the Carnegie Endowment, simply responded on Twitter, "Well, this sucks."
Initially the US attempted a mixture of bad cop and good cop tactics to try and convince the Russians to resume inspections. Initially they went stick, saying that if the Russians weren't going to publish data on deployed systems and warheads, they were going to stop doing so as well. Then after the Russians basically shrugged and said "Fine then," in May 2023 they tried something a little bit different by attempting to basically shame Moscow into compliance. At that time, even though the Russians were no longer reporting, the Americans decided to publish their figures anyway.
Declaring 662 deployed ICBMs, SLBMs and heavy bombers, 1,419 warheads for them and 800 deployed and non-deployed launchers. Russia didn't reciprocate, and about a month later in June 2023 the Americans revoked the visas for Russia's nuclear inspectors. From that point verification for both sides probably became more of an intelligence problem than a treaty problem.
And while the war in Ukraine has arguably served as a catalyst for breaking down this existing nuclear treaty system, it's also caused a lot of the objections both sides have long had to that treaty regime to bubble up to the surface. One major objection to just carrying on with New START that's been raised in both Moscow and Washington for example is that it only covers the US and Russia. New START is fundamentally a continuation of bilateral agreements between the US and Russia, and before that the US and Soviet Union, dating back to the 1970s. The schemes were originally designed at a time when the US and the Soviet Union had enough nuclear firepower to destroy human civilization several times over, and everyone else was a bit of an also ran.
And so the goal of those treaty arrangements was just to dial down the superpower numbers. Now that the numbers have been dialled back however, both sides have reasons to want to bring other powers into the treaty tent. Some Russians have complained for example that New START places absolutely no restrictions on the French and British nuclear arsenals. That's still not hugely significant numerically if you count warhead inventories as a whole, including those that are in reserve or awaiting disassembly.
But if you're primarily concerned with warheads that might be actively deployed on submarine-launched ballistic missiles, adding the French and the British to the equation might be enough to boost the US numbers by a double digit percentage. Not to mention they give Europe access to its own nuclear deterrent, which is probably an inconvenience for Russian foreign policy. On the other hand, some American strategists have complained that the treaty arrangements tie their hands somewhat, but do nothing to restrain the People's Republic of China. As we covered in a previous episode, for years now China has been in the middle of a massive build up in its nuclear forces with new ballistic submarines, new missile designs, and hundreds upon hundreds of additional land-based missile silos. And so, the argument sometimes goes, if Russian and US delivery system and warhead numbers continue to be capped but China's aren't, there might actually be an incentive on Beijing to try and put itself in a position of nuclear advantage, believing that might be achievable without causing an arms race because America was treaty bound not to follow. I think it's notable that US General Anthony Cotton, the head of Strategic Command, when talking about his hope for future negotiations actually ended up listing China before Russia.
"Would I love to see China step up and want to have a negotiation with us, would I love to see Russia come back? Absolutely. But I'm also a realist who understands that that may or may not happen. So as a combatant commander my job is to understand how do I build a force that I could present to the President if that doesn't happen." Meanwhile it seems that Beijing, London, Paris and others would be perfectly happy for Russia and the United States to agree an additional treaty limiting the US and Russia. Whether that's going to fly in the new security environment of the 2020s remains to be seen. What we have been seeing recently though, probably doesn't give immense reasons for optimism.
As we'll discuss a little bit more later on, despite the enormous resource demands of the war in Ukraine and the de-prioritisation of entire parts of the Russian armed forces, including the navy, Russia appears to have continued dedicating scarce resources to nuclear modernisation and the development of a range of "super weapons". They also recently took the step of testing the new Oreshnik intermediate range ballistic missile equipped with multiple independent re-entry vehicles by firing it against the Ukrainian city of Dnipro. And while we talked about the political and signalling implications of that missile last week, from a technical perspective the MIRVs themselves were interesting. Yes, the Oreshnik appears to have been carrying inert MIRVs that did relatively little damage based on what we can see in the satellite imagery, but it sure was carrying a lot of them. The footage we have appears to show 6 groups of objects striking the city in sequence, each group containing 6 projectiles. You can see a still image on the right there.
One Ukrainian source I've seen suggests that was the result of 6 MIRVs being used, each equipped with 5 additional sub-munitions or penetration aids. Missile tests, even against an opposing country, are expensive because missiles are expensive. And you are only going to be able to have so many of them, even as a relatively well-funded force. And I'll leave it to the audience whether or not Russia qualifies as such at the moment. Essentially what I'm saying is Russia probably made a very considered decision to fire the Oreshnik with this particular warhead configuration, understanding full well what that would signal to the world, what data it would provide them, and what concepts it would validate. And if you were trying to go back to a treaty arrangement which encouraged everyone to optimise their missiles to deliver single large warheads rather than MIRVs, it perhaps would have been an interesting choice to test and demonstrate your capacity to deploy 36 objects off a single intermediate range missile.
Meanwhile in the US, which only finished un-MIRVing its Minuteman III missiles in the 2010s, the momentum may now be swinging in favour of re-MIRVing them. In February of this year General Cotton reportedly told the Senate Armed Services Committee, "I do believe that we need to take serious consideration in seeing what uploading and re-MIRVing the ICBMs look like, and what does it take to potentially do that?" And just last month in November, the US would end up test firing a Minuteman III carrying three dummy warheads. It's worth noting this is far from the first time the United States has done this since the beginning of New START.
And they've generally sought to demonstrate that the Minuteman III remains entirely capable of carrying MIRVs, even as the US has transitioned away from deploying them. But against a backdrop of some of these statements, actions and general tensions in the treaty regime, there's always a chance a test like this will be taken as having signalling or demonstrative value. The US has also reportedly made leaps in other areas including during the warhead modernisation process.
For example it was claimed in an article published by the Bulletin of the Atomic Scientists that the deployment of a new arming, fusing and firing system on the warheads for US SLBMs had significantly increased their ability to destroy hardened targets, like missile silos. "The result of this fusing scheme is a significant increase in the probability that a warhead will explode close enough to destroy the target, even though the accuracy of the missile warhead system itself has not improved. As a consequence, the US submarine force today is much more capable than it was previously against hardened targets such as Russian ICBM silos. A decade ago only about 20% of US submarine warheads had hard target kill capability, today they all do." So basically even as US MIRVs may not have been increasing in numbers, they may have been increasing in counter-force capability.
Which I think naturally leads to the question of who the potential biggest losers are here of this particular Russian decision to start walking away from New START. I'd suggest there's an argument that everyone loses to at least some extent, but that doesn't mean the impact of a treaty system collapse is likely to be symmetrical on all nuclear powers. For Russia for example, I think a major question is can the country really afford a major nuclear build up at this time? Yes, an end to the treaty system might allow them more freedom in rolling out new nuclear delivery systems and moving a lot more reserve warheads back into deployed status. But that presumably has to be balanced against the increased freedom it would give the US nuclear forces, and also the question of whether or not the budget is really there for this kind of build up. The war in Ukraine is already functioning as a bit of a black hole for Russian military resources. So I think it's an open question just to what extent Russia really would divert resources away from things that are potentially useful fighting that war so that Moscow can instead threaten Western powers with 2,000 deployed warheads rather than 1,500.
And the US is arguably in a pretty good position to counter any perceived Russian build up in the short to medium term by simply moving warheads from reserve into deployed status. By most estimates the US has more warheads in reserve than it currently has deployed. The table on the right suggests around 1,900. And if the US reaches a point where they no longer consider themselves bound by those treaty limits, there is enormous capacity on the deployed delivery systems to take more warheads. The 400 Minuteman III missiles for example currently deploy with 400 warheads.
But the estimates on the right suggest you could probably double that. The submarine force too is also believed to have a lot of slack capacity for additional warheads. This is one of the reasons why Matt Korda, one of the co-authors of the Nuclear Notebook, said of the Russian decision, "This is a massive own goal by Putin, Russia benefits from New START just as much as the United States. This decision is clearly political and emotional, not strategic."
Another co-author of the Nuclear Notebook, Hans Kristensen, attempting to quantify the potential effect of the Russian decision, said of New START, "Without it, the United States could double its deployed arsenal." So in the short to medium term at least, even if Moscow believes they might be able to drive up American anxiety levels by suspending inspections, It's not really clear how they'd build up a significant advantage over US nuclear forces in the short to medium term especially given the resources Russia is likely to have available and the fact that America has a lot of reserve warheads. For the UK and France, any Russian nuclear build up is likely to be unwelcome, as is a decrease in the transparency around the nuclear treaty system.
But neither side really relies on things like ground-based ICBM silos that might be potentially vulnerable to a Russian counter-force attack using MIRVs or hypersonic glide vehicles. Instead they both rely on a continuous survivable at-sea nuclear deterrent. They both have programs in place to replace their existing generation of ballistic missile submarines, and sustain that survivable deterrent for decades to come. Plus with the French, to neutralise their nuclear deterrent you'd also have to destroy their carrier and carrier aviation, lest the nuclear warning-shot weapons available for the Rafale get converted into parting-shot weapons. So if we were to try and single out the types of countries that might be most disadvantaged by a collapse in this Russia/US treaty system, one nomination would probably be the People's Republic of China. As well, separately, as countries like Iran and North Korea that might consider themselves potentially at risk of a US counter-force attack.
As we've covered in previous videos, China is in the midst of a major nuclear weapons build up, and is believed to be pushing closer towards nuclear parity with the United States. From the Chinese perspective, an agreement like New START clearly fixed the marker of where parity could be achieved. No matter what nuclear forces the Chinese built up, it was reasonable to assume the US wouldn't go beyond 1,550 deployed warheads because that was the treaty limit. But if New START continues to break down and isn't renewed, the goal posts for China's nuclear build up might shift. The American 1,500 warhead force might become the American 2,000 warhead force in pretty short order. And from a budgetary and planning process it might make any Chinese build up towards nuclear parity with the United States much more of an uncertain process.
Where there would always be a non-zero chance that an increase in Chinese nuclear spending might be matched by an increase in US nuclear spending. So in that respect I'm sure Beijing is absolutely thrilled with all the decisions Russia has made regarding the New START treaty since 2022. The bottom line here is that if we recall those first principles we discussed earlier about when you might be likely to get a treaty and when you might not, obviously nothing is certain and there are always going to be incentives on countries to try and limit their collective spending on nuclear weapons. But it's been suggested there are at least some reasons to be a bit pessimistic about the future of existing restrictions on things like delivery systems and warhead count. That's both because restoring New START now and preventing it expiring entirely in early 2026 would require Washington and Moscow to be able to negotiate and reach an agreement in good faith.
And at the moment, as you may have noticed, there are a couple of reasons that might be difficult. The other is that the global strategic balance now looks very different than it did in the 1970s, or even in 2010. The US increasingly considers China, not Russia, to be its most powerful competitor. The US is therefore likely to want any future agreement not to weaken their relative position compared to Beijing, and want the Chinese to be part of any future agreement. Beijing's willingness to do so is of course far from certain. Russia meanwhile is likely both to have a range of new nuclear delivery systems that it's unlikely to want to see regulated in any future agreement.
And we've seen some Russian figures suggest that future agreement should cover the UK and France as well. Whether that's a recipe for a new robust agreement within the next 14 months is a question I'll leave to the audience. Which leads us to a brief closing discussion on the way global nuclear arsenals already appear to be changing, even though the US/Russia treaty system isn't technically dead yet. Russia for example is believed to have been investing very heavily in new families of delivery systems. One of the most significant
in a world where restrictions may eventually drop away is the famous RS-28 Sarmat. For Russia's competitors, one of the great challenges Sarmat poses if it's ever fielded in significant numbers and without attached warhead limits, is just that the missile has a massive payload. A US Minuteman III ICBM has a throw-weight payload of about 1.1 tons. Sarmat, according to the CSIS, is rated for about 10 tons.
That provides the potential to carry lots of relatively heavy MIRVs, hypersonic glide vehicles, decoys and penetration aids, or all kinds of things in between. You might fit 3 MIRVs on the Minuteman III, but up to 16 on the Sarmat. The consolation prize for other countries so far is that Sarmat, at least for now, appears to have a bit of a premature explosion problem. We believe the missile has been successfully tested exactly once, in April 2022. Contrast with the previous Soviet design that was extensively tested before going into service. The missile reportedly had a test failure in February 2023.
And blew up a not insignificant portion of the test site during a follow-up test in September 2024. But it would be deeply irresponsible to assume that Russia can't eventually fix the technical issues with the design, nor ignore the fact that we've just seen them demonstrat
2024-12-04 05:47
