Military analysts often use modeling to predict specific outcomes in war, including winners and losers, casualties, territorial gains or losses, and combat duration. But a potential U.S.-China war over Taiwan, likely also involving some American allies, poses analytical and policy challenges that make predicting outcomes especially difficult. In particular, the outcome of a Chinese maritime blockade of Taiwan scenario, in which a U.S.-led coalition aids Taiwan’s military to break the blockade and keep the island polity economically viable, may be too close to call.
In this paper, a combination of simple military modeling and path-dependent scenario or campaign analysis is used to determine whether the outcome of a maritime blockade of Taiwan can be feasibly predicted. The methodology draws from the well-structured and clearly described framework recently offered by Rachel Tecott and Andrew Halterman. Although I provide a limited analysis here, the results strongly suggest that any predictions by either adversary would be unreliable.
For this type of contingency, computations should seek to establish feasible upper and lower bounds related to the performance of each country’s respective command and control networks, submarine and anti-submarine warfare operations, missile attacks against key infrastructure and naval ships, missile defense systems, and other military systems. But, as the analysis in this paper demonstrates, one set of plausible modeling inputs, parameters, and assumptions could easily forecast a Chinese victory, while another comparably credible set could imply a U.S./Taiwan/allied victory. And this is even when accounting for a specific level of geographic and military escalation. As such, policymakers on both sides could not be certain that their own nation’s war plans would be successful. The dangers of escalation would add even more uncertainty to the situation.
With this in mind, the implications for force planning for the United States and its allies (and probably China as well) are discussed here. What becomes crystal clear is that both sides should avoid this type of war, now and in the future.
Acknowledgements and disclosures
Lori Merritt edited this paper and Chris Krupinski performed layout. I am grateful to many colleagues at Brookings as well as a working group on military modeling led by Caitlin Talmadge and Daryl Press, and to Barry Posen and Joshua Epstein who got me into this kind of modeling work in the first place.