I recently pointed out what I thought was the biggest news in the recently-released 2020 China Military Power report: an apparent more-than-doubling of the PLA Rocket Force’s DF-26 IRBM inventory.
This growth to IRBM launchers is a continuation of previous trends: the 2018 report had listed "16-30" launchers, then 80 in the 2019 report, and now 200 in this year's report. iiss.org/blogs/military…
I said in my commentary that, if this growth in capability is real, it could present a more significant challenge to the American way of war in the Western Pacific.
The report lists "200+" as the number of likely missiles available, given 200 launchers. We know from Chinese TV footage that DF-26 units practice reloading missiles, and that the missiles have different warhead types that are swappable.
Thus, if each launcher had only one reload missile available (and there may be more than that), this would mean an IRBM force of more than 400 DF-26's, all configurable to anti-ship or land-attack missions (including nuclear, though that's not what really concerns me).
Some observers may not be too concerned about the deployment of a single such weapon system, but the scale of this change matters: going from dozens of missiles to hundreds is a quantitative change big enough to drive qualitative effects, esp. given the longer range of the DF-26.
First, at sea the number of missiles could broaden the PLARF's antiship mission from a "carrier-killer" focus to a generic "ship-killer" mission. China itself describes the DF-26 as capable against medium *and* large ships. globaltimes.cn/content/119694…
This change could matter in how it intersects with the USN's distributed lethality construct: an effort to distribute offensive combat power - partly given the threat of what were dozens of DF-21D ASBMs - away from carriers to smaller, more numerous ships. usni.org/magazines/proc…
With a much greater number of available ASBMs, these smaller warships - LCSs, DDGs, and especially logistics ships - could become "ASBM-worthy" as well. businessinsider.com/in-war-china-u…
The story is no better for USAF tactical aircraft & bombers based in-theater. With smaller numbers of long-range missiles, air operations might be possible from Guam or dispersed locations. But with hundreds of missiles available, this seems unlikely IMO. defensenews.com/digital-show-d…
While some might point to the heavy bombers as an answer to providing land & maritime strike (I agree, to a point), one wonders how long they would survive - or be able to find targets - without effective tac air support available from local airfields. reuters.com/investigates/s…
The next way the DF-26 force matters is through its much greater range, in particular the specific additional areas that it can strike. In the Philippine Sea, areas of relative sanctuary beyond the range of the DF-21 (1500km+) lie well within range of the DF-26 (4000km).
These areas have mattered in how US defense thinkers looked at the regional A2/AD challenge, positing the ability to operate forces reasonably safely outside the 1st island chain as a means to enable operations closer-in to defend our interests & allies. cimsec.org/tightening-the…
Further southwest, PRC thinkers have obsessed over the "Malacca dilemma", with China's vital imports vulnerable to chokepoint interception en route. With large numbers of DF-26s, the PLA may now have the ability to strike US/allied warships attempting to maintain such a blockade.
The same could now be true in the vital sea lanes leading from the Mid East to Asia and Europe, with coverage extending from PLARF bases in western China.
Now, some commentators have been incredulous that the PLA's IRBM force could have grown so quickly given the scale of expansion that would be required.
From my perspective, I doubt that the DoD would just make up these numbers, and TELs are not like mysterious underground WMD sites or an adversary's strategic intentions. They are distinct physical objects that can be counted from space.
Perhaps they're not fully integrated into effective combat units yet; but if that is the case it's still just a matter of time.
Other observers were already tracking an unprecedented expansion of the PLARF; seems like a continuation of that trend to me. popsci.com/story/blog-eas…
Some observers have also doubted China's ASBMs have the ability to strike moving targets at sea. But for the 2nd year in a row the report states flat out that the DF-26 "is capable" of conducting strikes against naval targets. This is pretty strong language for an intel report.
To be sure, as has been discussed by USN's leadership before, the range arcs of the PLA's ASBMs are not impenetrable, nor is the PLARF its first "A2/AD" challenge.
I'm still working on my thoughts about what to do about all of this, as to be sure I don't think there are any easy - or even very palatable - answers.
But my sense is that the trajectory that we are on as a nation is not keeping pace with the threat to our interests and allies.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
I’m excited to share that @ChinaMaritime has just released a short paper I co-wrote with Mike Dahm titled “Flooding the Zone: The Use of Civilian Landing Craft (LCTs) in PLA Amphibious Operations”.
This paper follows the surprising observation last year that the PLA appears to be using civilian “LCTs” - a civilian derivative of WWII landing craft - to go straight to the beach in practice landing exercises.
...first up, China's nuclear warhead totals. This year's report says that warhead production has slowed, with a total in the low 600s, but that the PLA is still on track to have 1000+ warheads by 2030.
Regarding China's early warning capability, we get a lot more specificity than I've seen before: that China now has IR warning satellites that can detect and warn of an incoming ICBM within minutes.
Last week saw the release of the 2025 China Military Power Report (CMPR). For those not familiar, this is DoD’s Congressionally-mandated unclassified update on the Chinese military. It’s an annual feast of open-source data.
So, here are few thoughts (Part 1) on the report:
I'll focus on those things that I thought were most noteworthy: significant changes from previous reports, reveals of information not previously seen in the open-source world, etc.
Where there is a point of comparison I'll have the new report on the left, older on the right.
The new report is fairly different: for one thing, it clocks in at 100 pages to the previous report's 182.
From the top, the preface differs: the 2024 report's reads mostly like an intel product, where there's more proclamation of administration policy in the new one.
PRC SEALIFT UPDATE: based on a review of satellite imagery & AIS data from recent landing exercises, it appears China is practicing using dual-use civil-military landing craft for direct beach assault.
Of note, while info is sparse, these ships appear to exist in large numbers.
To the details: for a few years now, we've seen what've been known as "deck cargo ships" being used in PLA military transport exercises.
This has been making the rounds—another article calling for U.S. conventional submarines—so I suppose it's time to weigh in. In this case, the article is written by someone personally knowledgeable of submarine operations—a U.S. submariner.
...the article seems to be missing consideration of two key factors that IMO continue to make the idea of U.S. (manned) diesel-electric submarines a bad idea: advances in autonomy, and the PLA Rocket Force.
First, autonomy: he explicitly calls for these diesel boats as cruise missile platforms (SSGs), making a point to separate them from the multi-mission capabilities of U.S. SSNs.
No need for ISR, ASW, or torpedo capability. Ok, great—this sounds like a perfect job for an XXLUUV.
She's pretty clear about how we got where we are: close to a US-PRC war, that the problem started with Taiwan.
She says the situation is not pre-ordained, and that the best way to avoid a war is to restrain US military posture in the region and tell Taiwan they may be on their own.
(I'd say she's half-right: it didn't have to be this way; but I'd place the blame elsewhere.)