Xi Doesn’t Have to Invade Taiwan. For Now, Patience Is Enough for Him

In January 2026, Beijing announced an investigation into General Zhang Youxia, the highest-ranking officer in the People’s Liberation Army (PLA) and a man long regarded as Xi Jinping’s closest military ally. Less than three months later, Xi met in Beijing with Kuomintang (KMT) Chairwoman Cheng Li-wun. These two events look unrelated: one concerns the interior of the Chinese military, the other Taiwan’s domestic politics. In this text I argue that they are two sides of the same strategy. Xi is consolidating power over the military while simultaneously investing in a political path to Taipei, keeping the military option in reserve in case that path closes. An additional brake is the lesson of recent wars: Russia in Ukraine and the United States in Iran have shown that military force no longer guarantees victory. Xi therefore has reason to first test whether political and economic instruments will suffice. The key date is 2028, when Taiwan will elect a president and a new parliament.

An Army Without Commanders

The January announcement was not another round of the anti-corruption campaign but its breakthrough. Zhang Youxia, vice chairman of the Central Military Commission (CMC), was placed under investigation together with Chief of the Joint Staff Liu Zhenli. Earlier, Li Shangfu (October 2023), Miao Hua (November 2024), and the second vice chairman He Weidong (October 2025) had already fallen from the commission. The seven-member commission was in practice reduced to two people: Xi and Zhang Shengmin, the former head of the military discipline apparatus.

The campaign is not dying down. In the summer the army’s press organ declared 2026 the decisive year of reckoning with corruption, and a new wave engulfed Zhang’s network of connections in the air force and the Western Theater Command—environments that had until then been relatively protected. At the same time, reconstruction is under way: in July Xi promoted two new generals, one of whom took charge of the military discipline apparatus. The new CMC is to be announced in the autumn of 2027, at the end of the current term.

This chronology is crucial for the entire text: the new PLA command will emerge only a few months before the Taiwanese elections in 2028.

Consolidation Before the Congress

The simplest explanation of the purges is also the most convincing: Xi is sorting out the command before the 21st Party Congress in 2027, which will select a new CMC whose composition he himself will decide. The language of commentaries in the military press is telling: the charge against Zhang and Liu concerned undermining the system of the “ultimate responsibility” of the chairman of the commission—that is, Xi’s personal power over the army. This is a political charge, not a financial one.

There is also a factional dimension: around Zhang Youxia and He Weidong, overlapping patronage networks had been forming for years, and successive waves strike precisely at them.

This does not mean, however, that the purge is a signal of “invasion soon.” Supporters of another interpretation, for example the authors of USNI, see in it the removal of the last resistance to Xi’s plans in the “decade of turmoil.” Jamestown analysts go further and suggest that Zhang’s schedule for training for joint operations diverged from the 2027 deadline set by Xi for the capability to invade. This is a credible hypothesis and increasingly well supported: successive analyses of commentaries in the army’s press organ point to a dispute over the pace of force buildup before 2027, and Jamestown in October links the purge to Xi’s dissatisfaction with senior officers’ assessments concerning operations against Taiwan. It remains, however, an interpretation: the ministry’s communiqué speaks of violations of discipline and law, some experts caution that no disclosed evidence need reflect the true cause, and most analysts see in this purge primarily a struggle for power. Even if the hypothesis is true, removing the brakes increases Xi’s freedom of choice but does not increase the army’s capabilities. The gap between desires and possibilities therefore grows.

It is also worth remembering that “2027” is a capability target for the army, not a date for a political decision.

Xi Before the Plenum: To Command or to Endure?

On 26 October the Fifth Plenum of the 20th Central Committee will convene in Beijing, the most important meeting before the Party Congress in 2027, at which Xi is to receive a fourth term. According to official announcements, the main theme of the deliberations will be the Party’s rigorous self-supervision, not foreign policy. In the Communist Party a plenum is not a ceremony but a measure of the leader’s strength, so Xi faces a simple calculation. He can arrive as the commander who puts the ranks in order and calmly rebuilds the army, proving his power through discipline and the purge. He can also seek confirmation in a “short, victorious conflict” that silences critics and closes disputes over his legacy. The first path is cheaper. Getting stuck in a conflict over Taiwan—or small incidents at sea or in the air, or something more—would be a greater risk for a leader heading toward a fourth term than any faction in the army, and the army has just lost the commanders who could have led such a conflict. Xi does not have to prove anything on the battlefield when he is proving it in the offices, and the Kuomintang is working on his behalf at the Taiwanese ballot boxes.

The Price: An Army Under Repair

An amphibious operation across the Taiwan Strait is the most difficult undertaking one can assign to an army: it requires efficient joint command, logistics, and trust between levels. The purges strike precisely at these elements. Zhang once headed the CMC’s Equipment Development Department, so the blow hits the heart of PLA modernization. Veterans of the 1979 war with Vietnam, the last conflict in which the PLA actually fought, have also fallen from the command. Some Western assessments say that the shock reduces the army’s combat capability, at least in the short term.

There is, however, a paradox that is worth naming outright. A weakened army need not be safer for the region. Some analysts warn that the loss of experienced, “realistic” commanders increases the risk of miscalculation, because decisions increasingly rest on Xi’s personal preferences rather than on professional military advice. The purge therefore reduces the probability of a planned invasion in the coming years, but may increase the risk of an accident or a crisis escalating beyond control. Experts add a temporal dimension: the threat from Beijing may be weaker in the short term but stronger in the longer perspective if Xi staffs the command with more compliant officers.

Investment in 2028: Cheng Li-wun’s Visit

Xi’s meeting with KMT Chairwoman Cheng Li-wun on 10 April was both a symbolic and a strategic event. Beijing has long talked mainly with Taiwanese actors who accept its thesis that Taiwan is part of China, avoiding official contacts with the governments of the Democratic Progressive Party (DPP). Xi offered cooperation based on the 1992 Consensus and opposition to “Taiwan independence.” The visit took place amid a dispute over special defense spending, and the DPP accused the KMT that blocking arms purchases was to be the coin of exchange for the meeting. No evidence of such a deal was presented, but the association itself is already a political cost for the KMT.

The most convincing interpretation is simpler and less conspiratorial. The KMT is pursuing its own goals: it wants to win the elections, and for that it needs the image of a party of peace. Cheng called the visit a “peace mission.” This authentic strategy nevertheless has effects that Beijing can exploit: it strengthens the narrative of dialogue with “reasonable” forces on the island, weakens the image of China as an aggressor, and helps isolate the government of Lai Ching-te, whom Beijing regards as a “separatist.” The KMT is not necessarily playing under Xi’s dictation, but its political logic and Beijing’s logic overlap on a certain stretch. It is an unintended beneficiary, not an agent.

The test will be the local elections on 28 November 2026. Four years ago the opposition (KMT and TPP) won 15 races for mayoral and county magistrate posts—the KMT alone 14—while the DPP won only five. For the DPP the challenge is that it is the ruling party, and voters increasingly expect checks on power after years of its rule. According to a TVBS poll, support for President Lai stood at 38% after two years in office, with satisfaction with governance itself at 31%. The campaign is under way: for the biggest prize, New Taipei, a close race is being fought, and in September the KMT and TPP broke off cooperation in Hsinchu County.

The stakes in 2028 will not be only the presidency. In the 113-member Legislative Yuan no party has an absolute majority: the KMT caucus has 54 seats (including allied independents), the DPP has 51, and the TPP 8. The opposition therefore has 62 votes, enough to block and amend government bills, including the defense spending plan, but not to impeach the president: the motion to impeach Lai Ching-te failed in May 2026 for lack of the required two-thirds majority. The elections will be held no later than January 2028—the current parliamentary term ends with the end of the year—and in accordance with previous practice the president will also be elected at the same time.

This picture favorable to Beijing nevertheless has cracks that it must take into account. First, Cheng’s visit provoked divisions within the party itself. Second, she will not necessarily be the KMT candidate in 2028. Taichung Mayor Lu Shiow-yen, regarded as one of the main candidates, visited the United States in March—she met among others with the director of the Washington AIT office and publicly supported a clearly higher special defense budget than the party leadership. Her profile, pragmatic and open to the United States, differs from Cheng’s line. Beijing therefore has no guarantee that a KMT victory means a victory for the pro-China line. This divergence also complicates the picture of “the KMT blocks armaments”: Lu supported higher spending, and on 8 May, after 162 days of deadlock and disputes inside the KMT, the opposition passed a special defense budget of NT$780 billion (about US$25 billion). That is 38% less than the government wanted (NT$1.25 trillion, about US$40 billion), but considerably more than the NT$380 billion that Cheng initially pushed. Washington expressed disappointment at the cuts and, as analysts assess, is building relations with various factions of the party, anticipating uncertainty after the next elections. This is precisely what creates the need for Plan B.

Who Can Be the KMT Candidate in 2028

The KMT nomination is not yet decided. Below are profiles of the four most frequently mentioned candidates.

Photo	Candidate	Line	Strengths and weaknesses
	Lu Shiow-yen, Taichung mayor	Moderate, open to the US	Experience and international image. Weakness: according to April commentaries she is losing momentum, and her line differs from the harder course of the party chair.
	Chiang Wan-an, Taipei mayor	Moderate, distances himself from Cheng’s line	Leads in some polls. Weakness: 47 years old and less experience than Lu; in November he defends his post in a difficult campaign.
	Han Kuo-yu, Speaker of the Legislative Yuan	According to commentators he positions himself as moderate	Recognition and parliamentary role. Weakness: loss in the 2020 presidential election and recall from the Kaohsiung mayoralty.
	Cheng Li-wun, KMT chairwoman	Harder, more pro-China	Controls the party apparatus. Weakness: fears that her views will lower the party’s electoral chances; no official declaration of candidacy.

The order of the rows does not imply a ranking. In my view the contest will be decided mainly between Lu and Chiang, but the outcome of the local elections on 28 November may change that.

Taiwan’s System in a Nutshell

Taiwan has a semi-presidential system: the president is directly elected, but the government can fall in parliament. Since 2024 the president and parliament have come from opposing camps, which explains many of today’s conflicts.

Since December 2025 a dispute over countersignature has been under way. For the first time in history the premier refused to countersign a law passed by parliament, and the opposition responded with an attempt to impeach the president, which failed in May 2026. The premier subsequently repeated such refusals many times. A no-confidence vote in the premier is risky for the opposition, because after its passage the president can dissolve parliament.

Plan B: When the Military Returns to the Game

Let us begin with what, with high probability, will not happen. An invasion of Taiwan before the 2028 elections is very unlikely, for four reasons. First: the command that would have to carry it out is shattered, and the new composition of the CMC will be known only in the autumn of 2027. Second: Beijing today has a more effective instrument than tanks and ships—the political channel through the KMT. Third: an amphibious operation requires trust in the chain of command, which the purges have just undermined. Fourth: the recent wars, in Ukraine and Iran, have shown the limits of military force, which Xi must take into account. There is one reservation, but a serious one: the risk is not a planned invasion but an accident or crisis that slips out of control when experienced commanders who tell Xi uncomfortable things are missing.

Plan B is not a deadline but a conditional scenario. Three variables can be distinguished in it which, in combination, would change Xi’s calculus.

The first: a step by Taipei toward formal independence. For Beijing this is a red line that is not only symbolic but legal: the 2005 Anti-Secession Law provides that in the event of Taiwan’s secession or the exhaustion of possibilities for peaceful unification, China may resort to non-peaceful means. Xi repeats it consistently, and after the May summit in Beijing President Trump reported that Xi again emphasized firm opposition to Taiwan independence.

The second: a change in US policy. Here the last months are ambiguous. After the summit Trump stated that he had not decided whether to proceed with a large arms package for Taiwan after hearing Xi’s concerns, and critics accused him that consultation with Beijing might violate the so-called Six Assurances of 1982. Since then the package worth US$14 billion has remained on hold. Trump described it as a bargaining card vis-à-vis Beijing, and the administration has not formally notified Congress of it, which is a condition for the sale to proceed, although it does not require prior congressional consent. Congress is not silent: a bipartisan group of senators demanded that the package be released even before the May summit, and at the end of September four Republican senators led by Mitch McConnell renewed the appeal. Xi’s September visit to Washington (23–25 September) brought no decision on the package, and Xi called on the Americans to stick to the line of opposition to Taiwan independence. The US ambassador to China assured after the summit that policy toward Taiwan had not changed. From Beijing’s point of view every signal of Washington’s hesitation lowers the costs of Plan B.

The third: a DPP victory in 2028, which would close the political channel. Beijing, which had invested in the KMT, would then lose its lever, and “peaceful unification” would be a less credible path, increasing the temptation of military pressure.

None of these variables alone is sufficient. Only their coincidence—a step toward independence, US hesitation, and a KMT defeat—could shift Xi’s calculus from pressure to force. In my assessment the first response would then probably be tools below the threshold of war: blockade, quarantine, economic pressure.

An Additional Argument: The Lesson of Ukraine and Iran

The fourth reason is more general in character. Ukrainian journalist Vitaliy Portnikov argues that military powers have ceased to rule the world. In his view the greatest asset of a superpower is the threat of using force, not its actual use, because with the threat one can obtain far more than with war. Two examples are supposed to be the proof. Russia has been waging war against Ukraine for nearly five years—it has not achieved its strategic goals and has not subordinated the post-Soviet space to itself, and in some respects, for example as a result of Finland and Sweden joining NATO, its strategic position has worsened. The United States, which on 28 February 2026 together with Israel attacked Iran, according to many assessments achieved its declared goals only to a small extent, whereas Iran gained real influence over traffic in the Strait of Hormuz, and its closure triggered a global energy crisis. Portnikov adds that Trump has already become convinced that the United States cannot dictate terms to China by force, because its economy depends on Chinese supplies.

The world therefore lives today only on the remnant of the conviction that military force can compel something, and Xi faces the same choice as Moscow and Washington. If he fails on Taiwan, he will strengthen the thesis that nothing is achieved by force and will undermine the myth of Chinese power. If he succeeds, he will break that rule and become, as Portnikov says, the only force capable of imposing its will. The stakes are therefore greater than the mere seizure of the island: Xi would be testing not only his own army but the entire rule that force still means something.

From this dilemma one can draw the conclusion that Xi has an additional reason for caution and for testing, before using force, whether political and economic instruments—the channel through the KMT, economic pressure, and isolating the Lai government—will produce an effect. The 2028 elections are for him a natural checkpoint: if this path works, force will prove unnecessary, and if it fails, its use will in Beijing’s eyes be a justified last resort. This is another argument that an invasion before the elections is very unlikely.

This logic nevertheless has limits that must be named. Xi chooses between temptation and risk. Beijing may yield to the temptation of exploiting US weakness, for example if conflicts in Europe, on the Korean Peninsula, and around Taiwan intensified simultaneously and the Americans lacked sufficient resources for a response. Assessments of the wars are also sometimes contested: Russia occupies part of Ukraine, and American strikes seriously destroyed Iranian military infrastructure, so lack of political success is not identical with military defeat. Finally, the lesson may work in both directions. Xi may conclude that the mistake of Moscow and Washington was not force itself but its poor selection and preparation, and that precisely this is what his purges and army modernization are meant to prevent.

What This Means for Europe

The European thread is secondary to the main thesis, but it is not of secondary importance. I see three dimensions.

US attention and resources: Every crisis in the Taiwan Strait, even below the threshold of war, shifts Washington’s attention and resources to the Pacific. For frontline states such as Poland this is not an abstraction: Europe must assume that American capacity for simultaneous support on two theaters is limited.

Dependence on China: A blockade or quarantine scenario would strike supply chains, including semiconductors on which European industry depends. The EU is debating de-risking relations with Beijing, but the lesson from Taiwan is that dependence must be reduced before a crisis arrives, not during it.

Analogies with Ukraine: Pressure on Taiwan and pressure on Ukraine share a common pattern: supporting the political option of concessions, crushing the opponent who does not yield, and testing the West’s determination. The analogy has limits, however, because Taiwan’s situation, its legal status, and the character of the threat differ from Ukraine’s. It is worth using as an analytical tool, not as an identity.

Conclusions and Indicators to Watch

The most probable scenario for the next two years is political and hybrid pressure, not war. Xi is rebuilding the command, investing in the KMT, and keeping the military option in reserve. Added to this is the lesson of Ukraine and Iran of which Portnikov speaks: since military force increasingly rarely brings victory, Xi has reason to exhaust the possibilities of politics and economics before using it. The risk lies not in a planned invasion but in miscalculation or an incident. Three groups of indicators will help read the situation:

Personnel: the results of the plenum of 26–29 October 2026, the composition of the new CMC in 2027, promotions of officers with experience in joint operations versus promotions of the political and discipline apparatus, further purges in the air force and the Western Theater Command.

Political: the result of the local elections on 28 November 2026 on the island of Taiwan, the course of the KMT presidential nomination and whether a more pro-China line or a moderate, US-open line wins, any change in Beijing’s tone toward the KMT, Washington’s decisions on the arms package and further Trump–Xi meetings, still announced for this year.

Military: the scale and character of maneuvers around the island, amphibious and logistical exercises, possible attempts at blockade or quarantine, incidents in the strait.