Who's Actually Ahead? New Reads on the PLA Balance

Opening

Welcome to today's China report. Today it's the People's Liberation Army, the PLA, and the US-China military balance. Four assessments landed within a few weeks of each other, and they don't agree on the one question that matters: whether the balance is still sliding toward Beijing or has started to settle. We've got the Center for Strategic and International Studies on Chinese aircraft carriers, Brookings on the balance domain by domain, and the Lowy Institute in Sydney on what all of it means for Australia. And one that isn't squarely on topic: a report from the Center for a New American Security about Russia, which earns its place because it supplies the allies-and-partners variable the other three hold constant, so don't ask it questions about the PLA it never set out to answer.

CSIS on holding China's carriers at risk

First, "Flipping the Script", published on the third of August by CSIS, by Benjamin Jensen, Mackenzie Eaglen and colleagues. The argument is a reversal. For twenty years the anti-access debate has run one way, how does China sink American carriers. This report asks the mirror question. China now has the Fujian, an eighty-thousand-ton conventional carrier, the Sichuan, a forty-thousand-ton amphibious assault ship, a nuclear-powered hull under construction, and a Pentagon-sourced trajectory toward nine carriers by twenty thirty-five. Those are big, visible, slow to replace. So hold them at risk: submarines forward, land-based anti-ship missile batteries strung from Japan to the Philippines, long-range air and space-enabled fires, cyber and electromagnetic attack, and cheap unmanned swarms plus strikes on ports.

The mechanism is that kill chains are reciprocal. If Chinese sensors, escorts and logistics can find and finish an American carrier, then the same physics run backwards. And the goal isn't sinking anything. It's making routine operations, training, deployment, showing the flag, expensive enough that the carriers stay tied up.

Here's what it rests on. There's no original modelling. It synthesises the Pentagon's twenty twenty-five China report, CSIS's own missile and China Power datasets, Congressional Research Service work, a twenty twenty-three Chinese academic study running twenty simulated battles, and a British Ministry of Defence assessment. The one number doing real work is a production figure: American Mark forty-eight torpedo output runs about seventy-nine to a hundred and twenty a year, at roughly four point one million dollars each, against an estimated wartime demand of sixty to a hundred and twenty a week. Read that again. The report's own arithmetic says the strategy it recommends cannot currently be supplied.

Which I think is the honest, valuable part. But here's the weakness. Elsewhere in the same report, defending American carriers, the authors argue that actually killing a carrier is far more complex and uncertain than people portray. That's a good point, and they don't apply it to themselves. And China's carriers aren't where PLA striking power over Taiwan lives; the rocket force and shore-based aviation are. This may be cost imposition against the least important target. What would settle it is a cost-exchange calculation and sustained tracking data on carriers running quiet. The report gives neither.

Brookings on who holds the advantage

Next, "Who holds the advantage?", published fourteenth July by Brookings, three assessments, from John Culver, Bruce Jones and Michael O'Hanlon. Culver says America keeps undersea warfare, carrier aviation, bases and allies, while China leads in precision missiles and, durably, in the scale and modernity of its defence industry. Jones says inside the first island chain China has the upper hand in surface warfare and air defence, while America holds undersea and space, and notes China is already outperforming the US on large multipurpose undersea drones. O'Hanlon says flatly that China is not ahead and not close to parity, though he concedes a blockade would be too close to call.

The shared move is that aggregate scorecards are the wrong instrument, because everything is scenario-dependent. Fine. But that only works if domain advantages don't cascade, if losing air defence inside the first island chain doesn't spend your undersea edge. Nobody defends that assumption.

Evidence: this is expert judgement over public sources, the Pentagon's China report, Hudson on air base hardening, the International Institute for Strategic Studies on submarine production. No dataset, no wargame, no original modelling. Numbers are sparse. Culver's sharpest is that Xi has removed at least fifty-three percent of three- and four-star officers since twenty twenty-four, and that the newest Los Angeles-class submarine was commissioned in nineteen ninety-six. O'Hanlon offers that China's navy has more ships but around half the tonnage.

It's labelled as judgement, and that's fair. But three men reading the same Pentagon report and the same CSIS analysis isn't three independent lines of evidence, it's one evidence base and three priors. The disagreement is about weighting, not facts. And because it's curated monologues with no cross-examination, O'Hanlon's "not close to parity" never has to meet Culver's industrial-base claim head on. None of them says what would change their mind.

Lowy on the threat to Australia

Third, the Lowy Institute, fourteenth June, Sam Roggeveen and David Vallance. Their claim: American primacy in the Indo-Pacific is already gone, and the near-term threat to Australia isn't missiles on the mainland. It's trade interdiction, undersea cable cutting, cyber operations and naval presence, ranked, in that order, as things that are possible now.

The method is deliberately capability-based. They say explicitly they're not assessing intent, because you can't plan against intentions. So they project force structure to twenty thirty-five: twenty-five nuclear attack submarines, thirty-five conventional boats, nine carriers, a marine corps of eleven brigades.

The number carrying the argument is Australian, not Chinese: ninety-nine percent of Australia's international trade by volume, one point six billion tonnes, six hundred and fifty billion Australian dollars a year, moves through ports. You don't need to strike Australia. You need to touch shipping. That's strong. So is the Australian Signals Directorate's own confirmation of pre-positioned Chinese access in critical infrastructure, and the observed trend of carriers spending fifty-eight days beyond the second island chain in twenty twenty-five against thirty-two the year before. Weaker: the twenty thirty-five order of battle is largely Pentagon projections restated, and the defence spending line, four hundred and seventy-four billion dollars in twenty twenty-four rising to nine hundred and seventy-seven billion by twenty thirty-five, is trend extrapolation presented as arithmetic.

It's the most candid of the four about its own limits. But it excludes logistics and command and control, and that's not a small exclusion when your headline threat is sustained interdiction far from home. Twenty-five submarines on a list is not patrols off Fremantle. Watch replenishment ship numbers, the report tracks deployment days and not the tankers.

CNAS on Russia and a Taiwan crisis

Last, "Forecasting the Future of the Axis of Upheaval", published fourteenth July by CNAS, by Andrea Kendall-Taylor. Written from Moscow's chair: Russia's partnerships with China, Iran and North Korea are opportunistic and bounded, not an alliance. In a Taiwan crisis Russia probably does more than it did for Iran, but not combat.

The argument is a ladder ordered by likelihood, UN vetoes, energy overland, sanctions evasion, intelligence sharing, cyber, then weapons and advisers, plus diversion in Europe and joint patrols near Japan. Direct naval involvement is rated highly unlikely, because Russia's blue-water navy is small, ageing and busy. And the ceiling is set by Beijing, which wants escalation managed.

The evidence is the June twenty twenty-five Iran conflict treated as a natural experiment: Russia supplied intelligence, targeting data, imagery, jamming equipment and upgraded drones, and stayed out. Sourcing is open, press, Kremlin statements, think-tank work, expert interviews, and it's funded by US European Command's Russia Strategic Initiative, which is worth saying aloud. The detail that matters most for this episode: Russian transfers of submarine quieting technology to China. Every other report here rests American advantage on undersea warfare.

The weakness is generalising from one case. Iran is a poor analogue for Taiwan, different stakes, different leverage. And "probably" is doing a lot of unquantified work.

What to watch

So: Brookings has O'Hanlon saying China isn't close to parity, and Lowy says primacy is already gone. Same public evidence, opposite verdicts, the difference is where you're standing, and neither side's data settles it.

Three things would move me. American torpedo and anti-ship missile production rates, because CSIS's own numbers say the strategy isn't currently buildable. PLA Navy replenishment ships, not hulls counted. And any further Russian transfer of quieting technology, because that's the one thing on this list that could take the undersea advantage everyone else is banking on.