China’s Dangerous Decline | Foreign Affairs Otesanya David December 23, 2022

China’s Dangerous Decline | Foreign Affairs

China’s Dangerous Decline | Foreign Affairs


The last two months have been among the most momentous in recent Chinese history. First came the 20th Party Congress, which President Xi Jinping used to extirpate his few remaining rivals. Then, a few weeks later, the country erupted in the most widespread protests China has witnessed since the mass demonstrations in Tiananmen Square and elsewhere in 1989. And then, barely a week later, came the startling denouement: in a rare (if unacknowledged) concession, Beijing announced it was loosening some of the “zero COVID” policies that had driven so many angry people into the streets.

It has been a head-spinning season, even by the turbulent standards of contemporary China. But underneath the noise, the events all carried the same signal: that far from a rising behemoth, as it is often portrayed by the U.S. media and American leaders, China is teetering on the edge of a cliff. Ten years of Xi’s “reforms”—widely characterized in the West as successful power plays—have made the country frail and brittle, exacerbating its underlying problems while giving rise to new ones. Although a growing number of Western analysts—including Michael Beckley, Jude Blanchette, Hal Brands, Robert Kaplan, Susan Shirk, and Fareed Zakaria—have begun to highlight this reality, many American commentators, and most politicians (ranging from former Secretary of State Mike Pompeo to President Joe Biden), still frame the U.S.-Chinese contest in terms of Beijing’s ascent. And if they do acknowledge China’s mounting crises, they often cast them as either neutral or positive developments for the United States.

But the opposite is true. Far from good news, a weak, stagnant, or collapsing China would be even more dangerous than a thriving one—not just for the country itself, but for the world. Dealing with a failing China could therefore prove harder for the United States than coping with the alternative has been. If Washington hopes to do so successfully—or at least fend off the worst of the fallout—it needs to reorient its focus, and fast.

Washington’s record in dealing with rivals in decline is not auspicious, and coming up with a new policy to manage China’s descent will not be easy. To make matters worse, it is unclear whether the Biden administration has begun working on the problem. But that is no reason for despair. There are a number of changes, some relatively easy, the United States could make that would greatly improve its odds—especially if it starts making them soon.


For many years following the death of Mao Zedong, China was the world’s exceptional autocracy: the sole large authoritarian state that seemed to defy the laws of political and economic gravity. Starting in the late 1970s under Mao’s successor, Deng Xiaoping, China gradually opened its markets, distributed executive power, imposed internal checks, promoted internal debate, used data to make decisions, rewarded officials for good results, and pursued a generally nonthreatening foreign policy. These reforms allowed the country to avoid the dismal fate suffered by most repressive regimes—including the famine and instability that China itself had experienced during Mao’s long reign. Under Deng and his successors Jiang Zemin and Hu Jintao, China did not just dodge such problems; it thrived, growing its economy by an average of almost ten percent a year between 1978 and 2014 and lifting some 800 million people out of poverty (among many other accomplishments).

Since taking office in 2012, however, Xi, in his single-minded pursuit of personal power, has systematically dismantled just about every reform meant to block the rise of a new Mao—to prevent what Francis Fukuyama has called the “Bad Emperor” problem. Unfortunately for China, the reforms Xi has targeted were the same ones that had made it so successful in the intervening period. Over the last ten years, he has consolidated power in his own hands and eliminated bureaucratic incentives for truth-telling and achieving successful results, replacing them with a system that rewards just one thing: loyalty. Meanwhile, he has imposed draconian new security laws and a high-tech surveillance system, cracked down on dissent, crushed independent nongovernmental organizations (even those that align with his policies), cut China off from foreign ideas, and turned the western territory Xinjiang into a giant concentration camp for Muslim Uyghurs. And in the past year, he has also launched a war on China’s billionaires, pummeled its star tech firms, and increased the power and financing of the country’s inefficient and underperforming state-owned enterprises—starving private businesses of capital in the process.

The recent party congress was just the icing on this toxic cake. Xi used the event to humiliate Hu, his predecessor and the last Chinese leader to have been chosen by Deng. He also replaced Premier Li Keqiang and stacked the Politburo and its powerful Standing Committee with loyalist hacks (most of whom have security, not technocratic, backgrounds). More than a display of China’s grandeur, the event served to highlight its growing flaws. It was Xi’s coronation as China’s latest Bad Emperor.

The damage Xi has wreaked is already starting to show in a host of ways. China’s economy has cratered under his capricious interference and the weight of zero-COVID (more than 313 million people were recently under some form of lockdown). The days of 10 percent annual GDP growth are long gone; while the government projects China will hit 5.5 percent this year, many analysts think it will be lucky to reach half that figure. The value of the yuan recently hit a 14-year low, and retail sales, corporate profits, industrial output, and property investment are all way down. Meanwhile, unemployment has skyrocketed, hitting 20 percent among young people over the summer. An estimated 4.4 million small businesses were forced to shutter last year, and informal data (official statistics are not available) show the country is also suffering massive brain drain as tech lords, other billionaires, and middle-class professionals rush for the exits.

Xi has removed most rivals and proved a superlative bureaucratic knife-fighter.

And things are likely to get much worse. As China stalls, it is increasingly unlikely to overtake the United States as the world’s largest economy. Instead, with innovation and entrepreneurship stifled and productivity declining, China will find itself mired in the middle-income trap. Domestic living standards may flatline or fall. And smaller budgets and bureaucratic incompetence will make it harder for Beijing to deal with its many dangerous preexisting conditions: a rapidly aging population, a massive debt load, a severe shortage of natural resources (including energy and clean water), and a wildly overheated real estate sector, the failure of which could pull down the entire economy. (Chinese households have more than two-thirds of their savings invested in property.)

As the situation worsens and the promised “Chinese Dream” recedes, popular anger will likely continue to bubble over, as it did last month. Few China scholars predict a full-blown revolution; Beijing’s machinery of repression seems too effective for that. But dissent among China’s ruling class is more plausible, as Cai Xia, a former professor at the Central Party School of the Chinese Communist Party, has warned. It is true that Xi has removed most rivals and proved a superlative bureaucratic knife-fighter thus far. But his purges have punished and humiliated as many as five million officials. That is a lot of enemies for any ruler—even the most ruthless—to manage.

As he tries to do so, Xi will also face external problems on just about every front—again mostly of his own making. Having abandoned Deng’s dictum that China “hide its strength and bide its time,” he has instead sought confrontation. That has meant accelerating land grabs in the South and East China Seas, threatening Taiwan, using usurious loans tendered under the Belt and Road Initiative to grab control of foreign infrastructure, encouraging China’s envoys to engage in bullying “Wolf Warrior” diplomacy, and most recently, backing Russia in its illegal and unpopular war on Ukraine. The consequences have been predictable: around the world, Beijing’s public standing has fallen to near- or all-time lows, while states on China’s periphery have poured money into their militaries, crowded under Washington’s security umbrella, and embraced new security pacts such as the Quadrilateral Security Dialogue (which links Australia, India, Japan, and the United States) and AUKUS (a trilateral security pact between Australia, the United Kingdom, and the United States).

In the years ahead, China’s problems will continue to mount—and to make matters worse, they will probably catch Xi by surprise since, under his totalitarian system, lower-ranked officials are now punished for sending bad news up the chain. As Shirk, a former U.S. deputy assistant secretary of state and the author of Overreach: How China Derailed Its Peaceful Rise, puts it: “People won’t dare tell [Xi] the actual downsides and costs of his policies and the problems they’re creating.” Even vital government-to-government communications no longer get through, which sharply increases the risk of an accidental conflict. As Matthew Pottinger, who was a top China adviser to U.S. President Donald Trump, recently explained, “We came to the determination during the Trump Administration that messages we were sending through diplomatic channels were not reaching Xi. The Biden Administration has come to a similar conclusion.”


American hawks will be tempted to celebrate China’s struggles. But they should postpone the party, for a declining China could be much more dangerous than a booming one. Given U.S.-Chinese interdependence, a weaker Chinese economy—especially one burdened by the inevitable tsunami of infections that will come as Beijing relaxes its COVID rules—will mean a weaker U.S. economy. (Just consider the global problems Apple recently suffered when Foxconn’s Zhengzhou complex erupted in labor disputes.) Although some scholars posit that China tends to turn inward when struggling with problems at home, decline can and has had the opposite effect on other countries, making them more unpredictable and belligerent. Brands, for example, has pointed to Germany in the runup to World War I and to Japan’s decision to attack the United States in World War II to argue that “peril may emerge when a country that has been rising, eagerly anticipating its moment in the sun, peaks and begins to decline before its interests have been fulfilled.”

The risks are especially great when that country’s leader has staked his prestige on big promises he feels he must deliver, just as Xi has done. Increasingly eager to bolster his credibility—especially after the very public failure and embarrassing reversal of his zero-COVID policy—and unable to rely on economic growth for his legitimacy (as previous Chinese leaders have), he may turn to the other arrow in the dictator’s quiver: nationalism. If he does, the result will be a China that looks and acts more like a supersized North Korea: a cash-strapped, repressive regime that provokes and threatens its adversaries in order to extract concessions, burnish its pride, and distract its public.

The greatest danger, of course, would be a military move on Taiwan. The parallels here to Russian President Vladimir Putin and his calamitous war on Ukraine are chilling. As Blanchette has written, “an environment in which an all-powerful leader with a single-minded focus cannot hear uncomfortable truths is a recipe for disaster.” Yet that is just the kind of system that Xi has created.


A U.S.-Chinese policy that accounted for all these dangers would require making a number of shifts in Washington’s current approach. First, the United States should do everything it can to ensure its own model is as attractive as possible. As a failing China becomes less and less enticing to other countries, the United States must polish up its own appeal. A good way to start would be to address U.S. political dysfunction. But at the moment, the prospects of doing so, and restoring trust in American institutions, seem dim.

A more achievable goal would be to avoid responding to Chinese provocations in ways that betray American values. As the political scientist and former Biden administration official Jessica Chen Weiss argues, by doing things such as blocking Chinese media access and restricting Chinese visas, “the United States has drifted further from the principles of openness and nondiscrimination that have long been a comparative advantage.”

On a related note, U.S. politicians should quit antagonizing China for narrow domestic political purposes. Hinting that Washington seeks regime change in Beijing, as Trump aides did on numerous occasions, accomplished little more than increasing China’s insecurity. The same can be said for purely symbolic and inflammatory gestures such as then–House Speaker Nancy Pelosi’s trip to Taiwan in August.

The more nationalistic Beijing becomes, the harder Xi may try to pick a fight—so the United States should avoid giving him any extra ammunition. The tricky thing, of course, is that poking the dragon is a long-standing American tradition for a reason: it plays well at home. Changing this approach will not be easy, therefore, especially as the United States heads into a presidential election. But avoiding gratuitous provocation is not a sign of weakness or the same thing as appeasement. To make that clear, Washington should visibly enhance its ability to contain a flailing China, articulate clear redlines, and end the policy of “strategic ambiguity” on Taiwan (as the political scientist Richard Haass and others have advocated) while also reemphasizing that Washington would oppose any Taiwanese moves toward independence. The United States should send these latter messages quietly, however—in direct talks with Taipei and Beijing—to avoid issuing a public challenge to which Xi feels compelled to respond. To further shift his calculus, Washington should also bolster U.S. military assets in areas of possible confrontation, such as the Western Pacific, and should do all it can to make Taiwan a harder target (a long-overdue project that is finally underway).

Publicly dunking on a struggling rival may be tempting, but it will serve no one’s interests.

Of course, sending quiet messages requires a means through which to do so. So the Biden administration should reestablish a robust channel for crisis communications and reengage with China diplomatically in a way the United States has mostly avoided for the last six years—not to reward China for bad behavior, but to make sure the two governments can talk when they need to.

When it comes to the economy, the Biden administration deserves some credit for recent moves such as the passage of the CHIPS and Science Act and the establishment of export controls that will limit China’s access to semiconductors and the materials needed to make them. These steps should reduce the United States’ economic and strategic dependence and slow China’s military advancement without provoking it excessively. But Washington should think harder about the tradeoffs involved in continuing to decouple. Although such moves might help further insulate the U.S. economy, they arguably promote protectionism and could create problems for U.S. foreign policy by reducing Washington’s leverage and decreasing Beijing’s incentives to cooperate.

The difficulties involved in striking the right balance point to a final principle: as Washington reorients its China policy, it must be modest, in two important senses. First, if and when China starts visibly deteriorating, the United States must avoid the kind of triumphalism that accompanied the fall of the Soviet Union (despite President George H. W. Bush’s efforts to avoid humiliating the Soviet leader Mikhail Gorbachev). Publicly dunking on a struggling rival may be tempting, but it will serve no one’s interests. American politicians, despite their understandable eagerness to score points at home, must remember that as China declines, Xi’s political incentives to pick fights may grow—but so will his material incentives to cooperate, since Beijing will have much less money and attention to use for solving problems. Washington still needs Beijing’s cooperation on a host of issues, such as fighting climate change and preventing future pandemics, and it should make that cooperation as easy for Xi as possible. That means toning down the gratuitous tough talk. And, as Shirk suggests, it means giving “Xi reason to believe that if he were to moderate his policies, the United States would notice, acknowledge it, and reciprocate in ways that would be good for China.”

The second form of modesty involves remembering just how hard a problem a failing China will present—and how bad the United States has been at dealing with such cases in the past. Consider the U.S. record on North Korea. It is true that Washington has managed to prevent the worst-case scenarios: although there have been plenty of threats, a few minor skirmishes, and a lot of missile tests, the Kim dynasty has refrained from starting an actual war with anyone since 1950. But Washington has simultaneously failed to stop Pyongyang from immiserating its own people; exporting illegal narcotics, counterfeit dollars, and weapons; and most important, developing a substantial nuclear arsenal. That is not for lack of trying; U.S. presidents going back at least to Bill Clinton have spent huge amounts of time and effort attempting to avoid these outcomes. But they have all failed—which says something important about the difficulty of the challenge. Now remember that the population of China is about 54 times larger than that of North Korea and that China’s GDP is almost a thousand times bigger, and the scale of the problem comes into focus. Managing China’s decline will be a long, difficult process, with painful tradeoffs; in truth, there probably is no way to fully insulate the United States and the rest of the world from the pain it will inflict. But that is all the more reason why policymakers should start focusing on it now.



Source link

Write a comment