Axis or arrangement? China's alignments in the 2026 Indo-Pacific

Opening

Welcome to today's China report. Today: is the China-Russia alignment an actual bloc, or a pile of deals that happen to point the same way, and what's the Indo-Pacific doing about it. Five reports, from the Lowy Institute, the Center for Strategic and International Studies, and the Council on Foreign Relations. Two of them aren't squarely on topic, one's about American munitions, one's about Australian diplomacy, but they price and test the bloc question from outside, so they earn their place, and I won't ask them for claims they never made.

Lowy on authoritarian cooperation

Nick Bisley's long analysis for the Lowy Institute, published on the second of August, asks whether China, Russia, Iran and North Korea are building a new illiberal order. His answer: they're really cooperating, and it's really not an order.

The argument runs like this. The cooperation is reactive. It's driven by shared opposition to the American-led system, not by a shared project. So it scales up when pressure goes up, but it can't produce an alternative, because there's no organising centre and no blueprint. His evidence for the "no centre" half is absence, no four-way leaders' summit, no authoritarian Quad, everything bilateral. At this year's BRICS foreign ministers' meeting, disagreement over the Iran war blocked a joint statement.

The best number in the paper: sixty-one exercises involving at least two of those four countries between two thousand three and twenty twenty-one, and thirty-five in the three years from twenty twenty-two to twenty twenty-five. That's a real acceleration. But it's attributed to, quote, "a CSIS report", with no named publication. And that's the pattern here. Bisley does no original data collection. The BRICS and Shanghai Cooperation Organisation share-of-world-population figures come with no source at all.

So: sharp synthesis, honest about its own limits. My problem is the benchmark. He judges the alignment by whether it can build a replacement order, and nobody in it is trying to. Test it differently. Does joint activity keep climbing when the pressure supposedly driving it eases? That's the question, and this paper doesn't ask it.

CSIS on China-Russia flights near Japan

Riho Aizawa, writing for CSIS on the twenty-second of July, says the China-Russia military relationship around Japan isn't a scenario. It's Tuesday.

She builds a ladder. Joint bomber flights annually since twenty nineteen. Twice a year since twenty twenty-two. Mutual landings in twenty twenty-two, fighter escorts in twenty twenty-three, the first nuclear-capable H-6N in twenty twenty-four. Then the ninth of December, twenty twenty-five: two Russian Tu-95s and two Chinese H-6s flying more than six hundred miles in a straight line out into the Pacific off Shikoku, a track that, extended, points at Tokyo. Four days before that, the carrier Liaoning transited the Southwest Islands. Three days before, two radar lock-ons against Japanese aircraft, one lasting about thirty minutes.

The evidence is almost all Japanese Ministry of Defense press releases, dates, aircraft counts, coordinates, and the author is a fellow at Japan's National Institute for Defense Studies. For open-source military analysis, that's about as clean as it gets.

The conclusion is where I'd push. She notes that after a big February show of force, four American B-52s and eleven Japanese fighters, no comparable joint bomber flight followed for weeks. It's offered gently. But it's the load-bearing observation under a recommendation to add compellence to deterrence. One quiet fortnight, no control, and no count of the flights that didn't happen for boring reasons like weather or airframe availability. That's a lot to hang on an absence.

CSIS on China and North Korea

Victor Cha, Bonny Lin, Truly Tinsley and Andy Lim, same institution, sixth of August, on China and North Korea. This is the shortest thing in the set, call it three thousand words, and it argues Xi's June visit to Pyongyang bought a broad reset, driven substantially by Beijing losing ground to Moscow.

Their method is comparison against a baseline, and it works. In twenty nineteen, the summit readout promised a positive role in denuclearisation. At the United Nations Security Council in twenty twenty-three, China's ambassador said it out loud. In twenty twenty-six, the word simply isn't there. Same move on the military side. Defence Minister Dong Jun met his North Korean counterpart in June, first such meeting since nineteen ninety-two. A People's Liberation Army delegation went to Pyongyang last October, the first in six years.

Trade comes from Chinese customs data: two point seven billion dollars two-way in twenty twenty-five, the highest since the border closed in January twenty twenty, and the first half of this year running nineteen percent ahead. But still only about half of pre-sanctions levels. That last clause is the honest one, and I'd have led with it.

Here's my hesitation. The description is well sourced. The motive isn't. The claim that this is about offsetting Russia rests on a chart of high-level exchanges with no stated source, and in a piece this short, that's the load-bearing exhibit. What would settle it is whether the military channel produces an exercise or a transfer, rather than another meeting.

CFR on munitions and deterrence

Out of Ammo, from the Council on Foreign Relations on the seventeenth of August, by Rush Doshi and four co-authors. The claim: America has neither the munitions nor the factories to deter China over the next two years, so buy cheap. Drones, sensors, things you can spend freely. An eighty-five percent solution today beats a hundred percent solution that never arrives.

The logic is tight where it's about acquisition. High-end platforms can't scale inside two years, production timelines, materials, workforce, so the only capacity you can actually add in that window is cheap and fast. Twelve recommendations follow, and most are plumbing: schedule-driven acquisition authorities, multiyear contracts, pre-funding long-lead items like carbon fibre and energetics precursors.

On money, it's excellent. Of a hundred and fifty billion dollars appropriated to the Pentagon, at least seven point seven billion is explicitly for near-term precise mass, and over ninety percent of that was still unspent when they wrote. One programme: three hundred thousand short-range drones at two thousand three hundred dollars each.

But for a report called Out of Ammo, there's no ammo in it. No inventory baselines. No expenditure rates. No named wargame, no model, no Chinese production figures. The two-year clock rests on a stated Chinese ambition to be capable by twenty twenty-seven, not on an assessment of when they will be. And the Ukraine precedent gets one sentence with no numbers. This is a first-rate audit of American appropriations wearing the clothes of a threat assessment.

Lowy on Australia and China

Graham Fletcher, for Lowy on the ninth of August, on Australia and China. Fletcher was Australia's ambassador in Beijing. His verdict: a modest, pragmatic new normal built on self-interest, not warmth. Not a reset. Good enough.

The chain. A stable period, the free trade agreement in force in twenty fifteen, two-way trade up a hundred and twenty-five percent over the following decade. Four tipping points in twenty seventeen. Then coercion from April twenty twenty, coal, wine, barley, timber, lobster, cotton, aimed at a country sending about a third of its exports to China. Then stabilisation from twenty twenty-two, with the last restriction, on lobster, lifted in December twenty twenty-four.

And here's the hinge. Fletcher says the bans had, quote, "negligible macro-economic impact" and failed to move Canberra. If coercion failed, stabilisation isn't capitulation.

That sentence carries the whole paper, and it's asserted, not shown. No figures on where the trade went or what margin was lost. In fact the paper's trade numbers have no attribution anywhere, no statistics agency, no customs data, no polling. This is testimony. And an insider assessing a policy he helped make is the weakest available evidence on whether it worked. Two things I'll grant him. He's explicit about what stays frozen: the security dialogue hasn't met since twenty seventeen, the human rights dialogue since twenty fourteen. And he ranks his own risks honestly, putting a slow pile-up of disagreements above a Taiwan crisis.

What to watch

The real disagreement is between Bisley and Aizawa. He says the alignment is leaderless and won't shift the order any time soon. She says its China-Russia core is applying operational pressure right now. Both can be true, leaderless isn't harmless, but they'd fund different things. And note that Bisley's strongest number is borrowed from CSIS, so the two houses partly share a source.

Three things would move me. Whether China-North Korea military talk turns into an actual exercise or transfer. Whether that seven point seven billion dollars gets on contract by the end of this year. And whether the joint bomber flights near Japan hold their tempo through the rest of twenty twenty-six.