Thread by @tshugart3

Now that Balloongate has died down, a few thoughts on @CSIS’s recent Taiwan wargaming report. There was a fair bit of discussion of this a few weeks ago, but it’s taken me some time to review the report and gather my thoughts, so here they are:
csis.org/analysis/first…

There were a fair number of headlines about this report/study, some of them with fairly sanguine takeaways: bloomberg.com/news/articles/…

And indeed the overall finding was that China would be unlikely to succeed in an invasion of Taiwan, if (and this is a key if), 4 conditions hold:

These 4 conditions are non-trivial: much has been written about the sometimes-parlous state of Taiwanese military training and manning, for reasons not within the US control JPN might restrict use of bases on its territory, & the US may well not have enough long-range ASCMs.

Before we even get to those findings, though, with any wargame the entering assumptions, method of play, and player constraints of the game can strongly affect any such outcome.

After taking a close look at how the CSIS team set this up, I have to say I was deeply impressed by the thoroughness with which they considered just those sort of variables. IMO the vast majority of their (often tough) choices were entirely reasonable, and the process solid.

I particularly like how they made this an iterative process, running the game 24(!) times and twisting various knobs along the way. This is a great way to do it - seeing what works for each side or doesn’t, which variables/assumptions matter most, and what results are consistent.

Going into some of the game series’ assumptions, it won’t surprise my followers to see that I was quite happy to see the use of Chinese civilian RoRo shipping integrated into the analysis, as an additional source of cross-Strait sealift capacity, as a given.

This matters a great deal, as one of the main takeaways from the game was that a major limiting factor for success in an invasion was China’s capacity to land and sustain enough troops to take the island.

Since the survival (or not) of China’s invasion fleet is so important, one of the key variables the team looked at was the number of available anti-ship missiles. A key question they had was whetherthe new JASSM-ER long-range missile will have a maritime strike capability.

The team elected to use as a base case assumption that it would have such a capability, and as a pessimistic excursion that it would not. This mattered a lot in the game’s results, given the quite limited LRASM inventory (a few days’ worth), and much larger JASSM-ER numbers.

While the team’s decision to make JASSM-ER anti-ship-capable in the base case was a reasonable one, it’s an area where I differ.
While the USAF’s budget documents say nothing about an anti-ship version of JASSM-ER, it turns out the Navy’s FY23 docs do - it’s called the AGM-158C.

So I think the question of whether it’s likely that there will be a maritime-strike version of JASSM-ER looks to me to have been answered - it’s the Navy AGM-158C, and under current plans we’d have 31 or so available by 2026 (when the game is set).

For these reasons, I think it would make more sense to have it the other way round - as a base case that only LRASM and the 30ish Navy C-versions have anti-ship capability, and an excursion case where levers get pulled and that gets added to the USAF’s JASSM-ERs, too.

The other area where I differed with the CSIS team was in that question of sealift capacity. While, again, I was happy to see them take into account supplemental civilian sealift, IMO they somewhat underestimated the scale of that capacity.

For the 2026 timeframe, the report states that they anticipated an amphibious fleet that included 30 large civilian RoRos, and that the combined PLA Navy/civilian fleet could put ashore about 8K personnel, or 16K every 3.5 days.

In this recent article I estimated that, absent other constraints, China’s sealift capacity could deliver about 60K troops in the first wave, and then 300K troops in about 10 days (or about 30K per day).
warontherocks.com/2022/10/mind-t…

Also, my current count is that there are about 80 RoRo ferries operated by ferry companies known to be associated with the PLA, though I don’t know how many of those would qualify as “large” for them. Almost all those in my count are larger than about 4,000 displacement tons.

Clearly, there’s some daylight between our analyses here. As I said in my article, other than capacity China “will face other challenges, such as port loading, conducting logistics over the shore, or a lack of joint warfighting effectiveness.” That may account for some of it.

I imagine the truth is probably somewhere in between. In any case, IMO a useful excursion could be to analyze what happens if China went on a building spree, and had a much larger sealift force (naval, civilian, or a mix). After all, China is the world’s largest shipbuilder.

Remember - the US entered WWII with essentially no specialized amphibious assault ships, and <3 years later we did this:

Moving past assumptions, let’s talk about some of the key takeaways. First, a topic that’s been of interest to me for years - PRC strikes on U.S. bases in Japan: cnas.org/publications/r…

The team’s consistent findings were that the Chinese-side players consistently struck not only U.S. bases in Japan, but JSDF bases as well. These strikes were necessary, they worked, and they were of a net benefit to the PRC operationally, even if they brought Japan into the war.

Massed long range missile strikes had major effects in the game: 90% percent of U.S. and allied/partner aircraft were destroyed on the ground, and typically the U.S. lost two forward-deployed CVNs in the initial days of the war.

On the topic of Marine Littoral Regiments (MLRs), the team found what I have suspected: that while these units are nice to have, they lack the sheer number of weapons to really move the needle in a large-scale conflict like this.

Another key takeaway, and one that shouldn’t surprise anyone who’s spent much time contemplating modern naval warfare - expect heavy losses at sea, on both side. Typical for the game: 2 US CVNs and 7-20 major US warships. For China 50 or so major surface ships.

This means we should expect to lose more personnel in a matter of weeks, or even days, than we lost in the many years of the “forever wars”. There will be no “golden hour”, and in many cases recovery will not be possible at all. And there will be many MIA.

Ok, that’s enough for today. Next, I’ll take up the report’s many very solid recommendations. Until then…