Kamala Harris or Donald Trump, the US must avert a new nuclear arms race

An unpleasant reminder to Harris and Trump that their decision trees have many more branches than those facing presidents during the bipolar Cold War.

Right after moving into the Oval Office next January, either Kamala Harris or Donald Trump must open a dossier that voters prefer not to contemplate and candidates rarely dwell on. It’s the question of how to avoid nuclear Armageddon. And as during the early Cold War, the big decision is whether to enter (and try to win) a nuclear arms race against America’s authoritarian foes or to wrangle them into negotiating arms control.

The urgency springs out of a hairball of conundrums. One is the expiry, just a year after inauguration, of New START, the only remaining nuclear arms-control pact between the US and Russia. Its president, Vladimir Putin, suspended the treaty last year but has not yet violated its terms. That ambiguity is part of his Dr. Strangelove shtick since he invaded Ukraine in 2022 and started rattling his nuclear saber at the West.

Meanwhile, Putin’s buddies in Beijing are rapidly building out their own nuclear arsenal. China currently has about 500 atomic warheads and apparently wants to double that by 2030, then reach parity in the coming decade with the US and Russia. (The US has about 3,708 warheads, of which some 1,770 are ready for use on the tips of ballistic missiles or at bomber bases; Russia has similar numbers.)

More imminently, North Korean dictator Kim Jong Un — nowadays aligned with Moscow, Tehran and Beijing in hatred of Washington — could throw an “October surprise” into the US election season and test, for the seventh time since 2006, another fission bomb. Depending on what happens in the Middle East, Iran’s mullahs too could “break out” and make warheads. That would raise the number of countries with nukes to ten.

This dystopian mess is an unpleasant reminder to Harris and Trump that their decision trees have many more branches than those facing presidents during the bipolar Cold War. Even if you simplify the scenarios to just the two main antagonists, Washington confronts a vexing “three-body problem.”

The “no-limits friendship” between Russia and China raises the specter, as a congressional commission concluded, “of combined aggression” — that is, of them ganging up. So the next US president could infer that the American arsenal must deter not only the Kremlin or Zhongnanhai separately but also a joint attack. That could imply that the US needs to roughly double its nuclear arsenal.

If Washington were to start adding weapons, though, Beijing and Moscow would each become paranoid and accelerate their own arming efforts, forcing the US to increase its arsenal even faster. The question then becomes whether the US could “win” the ensuing arms race, and what such a victory would look like.

That starts with money, as banal as that may seem in the existential context of nuclear winter. Even without a new arms race, America’s upgrades of its nuclear arsenal are expected to cost more than $7.5 billion annually for years to come (currently more than 8% of the defense budget), with estimates constantly increasing.

Chronic delays and cost overruns account for the increases. For example, the Pentagon is trying to replace the aging Minuteman missiles with a new type called Sentinel. This month, the Pentagon said that the program is running 81% over initial projections and three years behind schedule. Similar snafus entangle programs for new submarines, bombers and warheads.

Now apply some multiple to these numbers to account for a new arms race against not one but two adversaries. The math looks ruinous. The cost would either come at the expense of America’s non-nuclear military prowess or of other economic ambitions. American society would be left weaker against its enemies than it could be otherwise. After all, those nukes are by definition unproductive — the idea behind deterrence is that the weapons will never be used and will end up on the scrap heap.

These costs would still be worth it if a bigger American arsenal were more effective at deterring. But is it? Some members of that congressional commission think so. Others, including its chair, Madelyn Creedon, protest that “when people say that ‘we have to have the same numbers as the combined total of everybody else,’ it makes no sense, it absolutely makes no sense.” The current White House also thinks that “we do not need to increase our nuclear forces to match or outnumber the combined total of our competitors to successfully deter them.”

Nukes are not like shells or bullets or drones, which enemy armies can keep firing at each other forever. Once any antagonist uses one nuclear bomb, opponents will retaliate, and the escalation spirals — with decisions made in split seconds — quickly become incalculable. That’s true whether the first strike is intended to be a relatively limited “tactical” nuke or a “strategic” knock-out blow.

In such an exchange, it only takes a handful of warheads reaching their destinations to erase civilizations and cause global fallout in the environment and food supplies. What matters for deterrence is not how many nukes the US or its adversaries have in total. It’s whether each is confident that some part of its own nuclear arsenal can survive an initial volley by the enemy (or enemies) and then destroy the other nations.

The US and Russia do have such a “second-strike” capability. For example, just one of the American submarines built for this purpose, of which 8 to 10 are at sea at any time, has enough missiles and warheads to annihilate a large country. China (which, unlike the US and Russia, has a policy of “no first use”) isn’t sure about its second-strike resilience yet. That, rather than a desire to intimidate the US, is probably the reason for its build-out.

The specter of an arms race reminds me of a metaphor that Carl Sagan, the late astronomer, sketched during the Cold War. “Imagine a room awash in gasoline, and there are two implacable enemies in that room: One of them has 9,000 matches, the other has 7,000 matches. Each of them is concerned about who’s ahead; who’s stronger.” The difference is that the room now has three or even more people in it. The similarity is that all of them would still be gone if anyone lit the first match.

This is the case for arms-control talks that the next president should make to Moscow and Beijing. That prospect may seem implausible now, with Putin waxing apocalyptic and Beijing suspending nuclear talks with Washington in protest over American arms sales to Taiwan. (The whole point of nuclear negotiations, as during the Cold War, is to bracket out this existential category of war from all other areas of conflict.) New leadership in the White House could change the dynamic, though.

It’s unclear how Trump, who mainly wants to punish China economically and has a simplistic notion of military “strength,” thinks about the problem. Robert O’Brien, his former National Security Advisor, urges his former (and perhaps future) boss “to maintain technical and numerical superiority to the combined Chinese and Russian nuclear stockpiles.” To that end, he even wants a second Trump administration to abandon the Comprehensive Nuclear Test Ban Treaty (which the US has signed but not ratified) and start blowing up prototypes again. And yet O’Brien recently told Bloomberg Opinion that “I don’t think Donald Trump is going to start a nuclear arms race.” That implies strategic confusion.

Harris, for her part, has had little occasion in her career so far to delve deeply into nuclear strategy. But her national-security adviser, Philip Gordon, has done just that: In the administration of Barack Obama he took an interest in nonproliferation (and helped negotiate the deal to keep Iran from getting nukes).

To her counterparts in Beijing and Moscow, a Harris administration would be harder to read at first but might come across as temperamentally less fissile. In this context, that seems advantageous.