Musings on missile defense
There are currently estimated to be 1588 deployed strategic nuclear warheads controlled by non-Western powers.
This represents an effective deterrent. Capable of destroying the war fighting capability of the collective West, killing hundreds of millions, and potentially plunging the entire world into a hard to reverse civilisational collapse.
Nuclear war is among the top anthropogenic x-risks and although rated by Ord as having a lower risk than some of the biggest x-risks such as misaligned ai or engineered pandemics, nuclear war is not a hypothetical risk. These weapons of immense power already exist in large numbers, and are possessed by hostile states, which, in the case of Russia, has employed the threat of their use to advance their strategic goals.
Nuclear weapons have already been used, twice, in the history of warfare, and there have been numerous documented ‘close calls’ that seemingly infer that the risk of a large nuclear exchange in the previous 60 years was high (above 10%). An equilibrium maintained by ‘close calls’ and ‘near misses’ will only last so long, until at some point a ‘near miss’ becomes a hit.
In the history of warfare new destructive weapon systems can alter the balance of power but invariably result in the development of powerful counter measures that blunt their effectiveness. The machine gun was countered by the trench, which was in-turned countered by the tank.
The pre-WW2 idea that ‘The bomber would always get through’ rendering other forms of warfare obsolete was invalidated by the deployment of high-performance fighters, the development of radar, and other forms of air-defences. Similarly, the initially hard to counter German U-boats which sunk an unprecedented amount of allied shipping in 1942 (seemingly rendering surface navies incapable of performing their historic task of protecting sea-lanes), were eventually countered by allied surface ships deploying new technologies (sonar, depth charges).
This pendulum of breakthrough and counter-breakthrough is the norm, and yet the US currently lacks a ballistic missile shield capable of blunting a major nuclear exchange, despite the threat of such an exchange being present for 60+ years.
This stands in contrast to other forms of missile, for which effective counter-measures do exist. Many armies & navies possess effective anti-missile systems, such as the Soviet era S-300 currently protecting Ukraine’s cities, or the Patriot SAM system. Israel’s iron dome missile defence system provided a very visible demonstration of this ability, claiming a 97% interception rate, against admittedly crude/outdated missile fired from Gaza.
The US does possess a number of platforms capable of intercepting ballistic missiles, but these are configured on the basis that its aim is to protect against small numbers of ballistic missiles from a rouge state.
America’s Ground-based Midcourse Defence (GMD) System has some 44 interceptors to protect the US mainland. Obviously, this leaves the US effectively defenceless against in a large nuclear-exchange. With Russia acting as a ‘rouge state’, the question is, should the US expand its ballistic missiles defence system to be able to counter or blunten a full-scale nuclear attack involving several hundred warheads?
Criticisms of missile defence generally hinge on two ideas:
- Missile defence is ineffective/unworkable
- Missile defence would undermine deterrence
This piece makes a good case against these ideas, auguring that: the US has already developed reasonably effective anti-ballistic missile systems (GMD, Aegis and THAAD), it’s simply a matter of procuring interceptors in sufficient quantities. Decoys are somewhat over-rated as a barrier, as advances in sensing technology make it easier to identify the true warheads amongst the decoys.
More broadly, it is worth considering that even if it is true that it is always more expensive to defend against ballistic missiles than it is to build them, we are not considering a situation in the case of Russia where we have two evenly matched peer competitors. Americas economy is around 14 times larger than Russia’s. Add up the GDP of the collective ‘Western’ alliance (US, Canada, Europe, UK, Japan, South Korea, Australia) and you get a bloc with a GDP around $50 trillion, outmatching Russia around 30 to one.
An effective anti-ballistic missile, the SM-3, has a unit cost of around $12 million. It is difficult to find unit costs for Russian ICBMS, but a Minuteman III missile costs around $7 million and has 3 warheads. Add the cost of the nuclear warheads themselves and you get a cost of $48.5 million each, or £16 million per warhead delivered. Russian missile and warhead costs will no doubt vary from this considerably, but it establishes a baseline that is surely right to within an order of magnitude.
Any anti-balistic missile system will need many times more interceptors than the aggressor needs warheads. But nuclear weapons are complex and expensive. When produced in great numbers, interceptor production can benefit from economies of scale.
How many times more interceptors are required than warheads? An SM-3 has an operational range of 1200km (although it is not clear if this is its range when employed in an anti-ballistic missile capacity). At this range around five launch sites would be sufficient to protect the entire continental US.
Having enough interceptors at every site to intercept every available warhead in Russia’s arsenal would require 7940 interceptors, meaning that if the interceptors had a 100% success rate (which they would not) each site would theoretically have enough interceptors to prevent an attack on any area of the continental US even if Russia’s entire arsenal was concentrated on that one area.
You could also reduce the number of interceptor sites to focus on the most highly populated areas. 3 interceptor sites could protect the overwhelming majority of the US population, leaving mostly lightly populated areas unprotected.
If you opt for a system that concentrates on protecting the over-whelming majority of population centres (as above) and stocks sufficient interceptors at each site to have 1.5 times the number of interceptors as potential warheads that could be launched at it (because interceptors do not have a 100% success rate), the system as a whole would require around 7100 interceptors. Such a system would cost around $85 billion for the interceptors (assuming cost per missile remains constant). Economies of scale in missile production could likely bring this figure down, but you would need to add the costs of additional missile tracking radars etc. $85 billion is clearly within the means of the US, given its $800+ billion defence budget.
This crude analysis is almost certainly wrong, but provides a useful context as to what’s possible.
Put another way: If we naively assume a missile + warhead cost of $16 million (taking the US figure and applying to Russia), and assume 6 interceptors are required for each warhead (2 interceptors per warhead to account for failures to intercept, divided across 3 sites covering most population centres), then the US would only have to outspend Russia 4.5 to one in order to maintain an effective anti-ballistic missile shield. The US already outspends Russia around 12 to one in terms of defence spending so spending 4.5x on missile defence what your opponent spends on nukes is not unreasonable. Countermeasures employed by Russia (e.g. better decoys) could be themselves be countered (better sensors to detect warheads vs decoys).
If missile defence focuses on cities/population centres, then it could help tip the balance in favour of a counter-force strike over a counter-value strike in the event of a nuclear exchange, which would drastically reduce the number of casualties.
It may once have been true that missile defence is ineffective/unworkable, but developments in anti-ballistic missile weapon systems means that is no longer the case. It remains true that the ratio of spend require to defend successfully would be considerably higher than that required to inflict damage, but the ratio itself is not so large that such a system would be outside of the means of the US against a weaker competitor such as Russia.
Leveraging US leadership in space
Austin Vernon has an excellent piece pointing out the incredible advantages Starship would provide the US when it comes to defence. Making access to space drastically cheaper could also unlock benefits for missile defence.
Space based interceptors could theoretically mean that missiles could be intercepted during their boost phase. As one missile carries several warheads (MIRVs), intercepting during the boost phase reduces the number of targets requiring interception several fold. A missile is also far easier to identify during its boost phase owing to the thermal radiation inevitability emitted by its rocket engine.
Such a system would be far harder to fool with decoys, as the decoys would have to be large and operational rockets, and not MIRVs. However, to be able to intercept a missile during its boost phase, you would need a huge number of space-based interceptors in a low orbit in order to ensure there would be an interceptor sufficiently close to the launch at any given time.
Theoretically, you could develop web of space-based interceptors forming a kind of ‘starlink of interceptors’ style constellation, but realistically the number required probably makes such a system unfeasible. It could be workable to counter small states like North Korea (a constellation of 496 systems would ensure at least three to four interceptors would always be within boost-phase intercept range of North Korea) but such a system would be easy to over-whelm with large volleys of ICBMs.
Directed energy weapons could potentially over-come this as each platform could engage multiple targets and could intercept targets during their midcourse phase as well as their boost phase, meaning they would have drastically larger operational ranges. This in turn means fewer nodes are required in the constellation. Clearly though, this requires hypothetical breakthroughs and could not be deployed in the short term.
Orbital mechanics means spaced based interceptors are probably presently unworkable for countering large numbers of ICBMs, but spaced based sensing systems could be an extremely useful component in a ballistic missile shield using ground-based interceptors. A constellation of just 24 missile tracking satellites would mean missiles launched from anywhere on Earth would be within view of sensors of at least two different satellites for the entirety of the midcourse phase of their flights. Such a tracking ability could be used to make interception by ground-based interceptors more accurate, as well as help distinguish warheads from decoys.
Wouldn’t a missile shield be de-stabilising by undermining MAD?
No ballistic missile shield will be perfect, and as pointed out here, even a 99% interception rate would mean 10–15 Russian warheads getting through. What US leader would launch a nuclear war on their opponent knowing that the response would result in several nuclear explosions on US soil with a US death toll potentially in the millions? (not to mention the many millions of civilian casualties wrought on the opponent).
Ballistic missile defence downgrades a nuclear exchange from a near-civilisation ending catastrophe to a horrendous travesty. Initiating a nuclear war would remain unthinkable for a democratic state.
It may also make an aggressor less likely to launch a nuclear war, knowing that doing so risks rendering their deterrence useless should the system prove effective. Better to keep a hypothetical sword of Damocles over your opponent’s head than to drop it and see it come to nothing.
Escalate to de-escalate
The Strategic Defence Initiative was a dud, but it did worry the soviets, and help apply pressure that led indirectly (literature is mixed) to the INF treaty. A large investment in missile defence could be used as a valuable bargaining chip designed to extract further arms reductions or otherwise apply pressure to the Russian regime.
Building a strong ballistic missile shield leaves Russia 3 options; 1) Accept a reduction in the effectiveness of its deterrence capability, rendering its threats and nuclear blackmail less effective. 2) Invest heavily in measures to over-come the system, requiring the diverting of scarce resources away from other military priorities. 3) Enter negotiations with the US that require the US to abandon the development of a ballistic missile shield in return for major reductions in the number of deployed warheads.
The US has effectively already incurred the costs of withdrawing from the anti-ballistic missile treaty. These being the Russian withdrawal from START II which would have eliminated MIRVs (Multiple independently targetable re-entry vehicles). So having incurred those costs, it is seems strange to not capitalise on the withdrawal to build a more significant missile defence capability. It may well be true that the world would be safer if we had START II and kept the anti-ballistic missile treaty, but we didn’t.
Perhaps the best way to realise further reductions in the number of deployed warheads is to invest in missile defence sufficiently heavily that it forces the Russians (or other nuclear weapon states) to negotiate in order to prevent it.
