Breakneck

China does the hard part first

Tyler Cowen called Dan Wang’s new book:

The best recent book on China, on China and America, and, arguably, the best book of the year flat out.

From the opening page, you can see what earned Breakneck such high praise. Wang starts off with a great example of how even an exaggerated claim can reveal a deeper truth:

Each time I see a headline announcing that officials from the United States and China are once more butting heads, I feel that the state of affairs is more than just tragic; it is comical, too, because I am sure that no two people are more alike than Americans and Chinese.

A strain of materialism, often crass, runs through both countries, sometimes producing veneration of successful entrepreneurs, sometimes creating displays of extraordinary tastelessness, overall contributing to a spirit of vigorous competition. Chinese and Americans are pragmatic: They have a get-it-done attitude that occasionally produces hurried work. Both countries are full of hustlers peddling shortcuts, especially to health and wealth. Their peoples have an appreciation for the technological sublime: the awe of grand projects pushing physical limits. American and Chinese elites are often uneasy with the political views of the broader populace. But masses and elites are united in the faith that theirs is a uniquely powerful nation that ought to throw its weight around if smaller countries don’t get in line.

In a recent interview, Tyler slightly dissented from this comparison:

What's striking to me about America and China is that when Chinese come here, they become Americans very quickly. You could say a lot of them are better Americans than the Americans by a lot of standards. When Americans go to China, they can never become Chinese. And that's the most fundamental difference. So if you look at the movement one way, the countries seem very similar. If you look at it the other way, they seem very different. And that's a pretty fundamental asymmetry. So I mostly agree with what Dan says in his book, but that's a big difference. They cannot swallow us. We can swallow them.

This is true, but I also suspect that it would be far easier for a Japanese immigrant to Shanghai to “become Chinese” than for a Chinese immigrant to Tokyo to “become Japanese”.

I also found Breakneck to be the best new book that I’ve read this year, although I don’t regard it as the best recent book on China. That honor goes to Peter Hessler’s Other Rivers, published last year. But it’s a close call—both are great books.

As with any good book on China, there’s a glass half full/half empty aspect to the analysis, with the first half presenting a relatively positive view of the advantages offered by an “engineering society” (as opposed to America’s lawyerly society), while the second half focuses on the dark side of turning your country over to technocratic engineers.

The book resonated with me for several reasons. First, the places Wang visits (Shanghai, Beijing, Tianjin, Guizhou, Chongqing and Yunnan) almost perfectly correlate with places that I have visited, often at roughly the same time. Second, the glass half full/half empty take is pretty similar to my own. Indeed, I’m a tiny bit envious that I didn’t write this book, although to do so I would have had to be more intelligent, better informed and a better writer than I actually am. It’s one of those “if I had wings I could fly” situations.

If I had to summarize the book in a single sentence it would be that China’s engineering society is better at building infrastructure, housing and manufactured goods, whereas America’s lawyerly society is better at encouraging artistic and scientific creativity and protecting human rights. My views on China closely align with Wang’s analysis, with one important reservation. I don’t quite buy his argument about manufacturing, presented in chapter 3. In the next few sections, I’ll make a few comments on each of the key chapters.

Chapter 2: Building Big

Over the past few decades, China has built physical capital at a pace far beyond anything ever achieved in human history. Chapter 2 focuses on housing and infrastructure, while chapter 3 focuses on manufacturing. Wang presents one superlative after another:

According to Viclav Smil, the 4.4 billion tons of cement that China produced from 2018 to 2019 nearly equals the amount of cement that the United States produced over the entire twentieth century

Read that again—from 2018 to 2019. And then think about the implications of the fact that China’s housing and infrastructure will soon be built out, and its population will be falling at an accelerating rate.

Beyond all the gaudy statistics, Wang also provides a sort of travelogue giving a ground level view of how China has been transformed by all this construction. In 2021, he and two of his friends cycled 400 miles from Guizhou (one of China’s poorest provinces) to Chongqing, which Wang calls “the embodiment of cyberpunk”.

In 2019, my wife and I visited Guizhou, and in 1994 and 2023 we visited Chongqing. I had many of the same impressions as Wang. Here's a subway that goes right through a Chongqing apartment building, from a post I did in 2023:

And here’s what I said in an earlier post entitled Land of Bridges and Tunnels:

Despite the attempt to rely more on tunnels than bridges, Guizhou ended up with 40 of the world’s 100 tallest bridges, including the very tallest. Read that again. I didn’t say China had 40% of the world’s tallest (which would be a major achievement), I said a poor, small province in the interior with only 2.5% of China’s population has 40 of the world’s 100 tallest bridges.

By the time Wang wrote his book, they’d built a few more:

Guizhou has built forty-five of the world’s one hundred highest bridges.

Many western economists focus on the enormous waste involved in building ghost cities and bridges to nowhere. Wang is more optimistic:

“Infrastructure investment can be too good for a country’s development level,” concluded a line from economist Michael Pettis, which was not an atypical sentiment. But China’s railways had been hugely crowded, with passenger trains sharing the same tracks as freight trains, which caused endless delays. The creation of this fast and dedicated passenger network relieved congestion for all.

A study undertaken by the World Bank in 2019 found that China’s high-speed rail system is economically viable, with ticket revenues able to recoup costs.

China is still just an upper middle-income country, but it won’t be for long. Because a new rail network is built to last for many decades (if not centuries), it made sense to build the system for the China of 2050, not the middle-income country that China was when the first line opened in 2008.

By focusing so much on the flaws in China’s huge infrastructure and housing build out, critics missed the glass half full aspect of this amazing undertaking:

People unable to appreciate the benefits of material improvements also don’t understand how it produces pride and satisfaction. China’s transformation has given people running water and toilets, mass transit and highways, beautiful parks and modern malls. Most people can remember a time in the recent past when they didn’t have those things. This growth trendline matters. The glittering skyscrapers and rail lines form a core plank of the Communist Party’s legitimacy. Though China’s growth has slowed substantially under Xi’s rule, people have hope for improvement. The better infrastructure that has been built helps people to feel that progress still courses throughout the country.

They didn’t have most of those things when I first visited China in 1994.

Suppose you were to ask some policy experts: “What’s the thing that the US has the most difficult time doing?” The answer might be building in densely populated areas, especially things like housing, adequate airports, freeways, high-speed rail and subways. I think of China as the country that did the hard part of economic development first and has yet to prove that they can do the less difficult stuff. Now it just needs to do the easy part—liberalize. Maybe not becoming as liberal as the West, but at least as liberal as Singapore.

The US still needs to do the hard part of building a great society. My living standards in Boston were materially harmed by the lack of a train stop in Newton Corner, where I lived two blocks from a commuter rail line. As a result, I usually drove to places like Harvard, downtown Boston, and Logan Airport, where parking was difficult. In a Chinese version of Boston, I would have taken an ultra-convenient subway to those places. Wang is certainly correct that Shanghai’s vastly superior infrastructure needs to be considered when comparing living standards with New York City.

In fairness, most Americans do not live in densely populated areas like NYC, and the infrastructure in places like Orange County is certainly adequate for my purposes. Wang is mostly speaking to an elite audience that lives in big coastal cities with substandard infrastructure.

I frequently argue that despite its name, the Chinese Communist Party is actually a right-wing organization. Wang has a similar view:

China is also a country governed by conservatives who masquerade as leftists. Perhaps no other self-proclaimed socialist country is as lightly taxed as China. . . . Low taxes make China stingy on welfare. Around 10 percent of GDP goes toward social spending, compared to 20 percent in the United States and 30 percent among the more generous European states.

China has replaced Marxism with nationalism, private businesses, social conservatism, a meager safety net and high rates of forced saving. I smile inside when commenters tell me that I’m a Marxist because I like some of the supply-side policies adopted by China. Words have truly lost their meaning.

Wang also presents the downside of all this investment spending, with a special focus on the Binhai area of Tianjin, which the local government tried to turn into another Manhattan. The problem is that while Tianjin is a very big city, it is somewhat in the shadow of Beijing. And the Binhai neighborhood is far from central Tianjin.

One downside of being married to an economist is that we have peculiar taste. Thus in 2019 I dragged my wife and daughter to Tianjin to inspect the Binhai development, which was mostly a ghost town. I even took a picture of a crazy library that was also discussed by Wang:

Here’s Wang:

The Dutch architects behind Tianjin’s Binhai Library put a bright white sphere in its center, around which undulating curves make up shelving space. Except few of these shelves held any books. Once I got up close, I could see that the beautiful [upper] shelves had only digital prints of book spines.

Many readers will view Wang as a relative optimist on China, although (as we’ll see) I’m not sure that’s entirely accurate. Indeed, in a few areas he may be presenting an overly pessimistic view. Thus, he emphasizes that China continues to be hit by serious floods, despite major flood control efforts. But unless I’m mistaken, the loss of life is no longer anywhere near as bad as in former times. And in a later chapter he discusses how bookstores prominently display books by Xi Jinping, but in Beijing (in 2024) I saw lots of books by people like Hayek, Mises, Habermas, Sontag, David Friedman and Brad DeLong prominently displayed on tables. Xi’s regime is increasingly repressive, but this is far from Maoist China.

Although the first three chapters of the book are mostly about China, Wang’s policy preferences seem aligned with the “abundance” movement in America. There’s even some strange new respect for Robert Moses. Wang suggests that we need to build more housing and more infrastructure, especially in the Northeast and California. I am in almost complete agreement with Wang’s policy agenda in chapter 2, and indeed throughout most of the book. The one exception is chapter 3, which covers manufacturing.

Chapter 3: Tech Power

China has what is by far the world’s largest manufacturing sector, which is moving into increasingly sophisticated sectors. Is this due to the efforts of the Chinese government, or did it occur despite government policies? And should we be worried that China manufactures far more than the US, both in aggregate and as a share of GDP?

I read Wang as having mixed feelings on the effectiveness of Chinese industrial policies and answering “yes” to the question of whether the US needs to do more to promote manufacturing. I’m more skeptical of industrial policy, and don’t worry about the size of manufacturing as a share of GDP, outside of a few sensitive military areas. (As an aside, I’ve long favored smaller fiscal deficits, which (unlike tariffs) would boost manufacturing.)

Given the size of China’s manufacturing sector, my argument seems counterintuitive. How could China be so dominant without any sort of affirmative government policies?

I’m old enough to recall when we went through the same debate about Japan, where planners (MITI) supposedly had magical powers that created Japan’s formidable export machine. This despite the fact that MITI told Honda to stick with motorcycles, and not to go into car-making. And despite the fact that Hong Kong grew even richer than Japan without any industrial policy.

Wang gives some credit to the Chinese government, but also suggests that the effectiveness of their policies has been overstated:

Then results of the Chinese government’s unceasing interventions in the economy are at best ambiguous. Economic studies have shown that the recipients of Chinese subsidies have, on average, lower productivity growth. . . . The forced technology transfer agreement meant to prop up China’s state-owned automakers instead robbed their need to invest in their own innovative capacities. China’s automotive successes come from companies like privately owned BYD, which had no foreign partners, after the entrance of wholly owned Tesla forced the company to raise its game.

Wang is referring to the fact that Tesla was not forced to share technology.

I suspect that most people underestimate the extent to which China’s manufacturing boom is a natural development, not an artificially created situation. My view is based on two facts:

1. China is exactly the sort of country that you’d expect to do well in manufacturing exports, even without any help from their government.

2. China’s huge aggregate output mostly reflects its huge population; in per capita terms China’s manufacturing output badly trails countries such as South Korea and Japan and is even further behind the US.

The world can be roughly divided up into two blocs. Countries in northern Europe (Switzerland to Scandinavia) and countries in East Asia have big trade surpluses and a comparative advantage in the export of manufactured goods. Most other countries run trade deficits.

I suspect this partly reflects cultural factors. These regions seem full of people who are well educated, have good attention to detail, and are willing to work with machines. In both Japan and Germany, I’ve seen tough looking young men obediently stand at a crosswalk waiting for the green light, despite not a car to be seen in any direction. That’s exactly where I’d want to manufacture cars!

Places like the Philippines, southern Italy, Greece and Portugal seem less amenable to manufacturing. Success in manufacturing also reflects that fact that Northern Europeans and East Asians are high savers, which leads to trade surpluses. China’s trade surplus is large in absolute terms, but not particularly large as a share of GDP.

Some economists react to this generalization by arguing that it isn’t just China, rather all of East Asia and Northern Europe engage in mercantilist policies and/or wage suppression to promote manufactured exports, but I see little evidence for that claim. Are places like Switzerland, Scandinavia and Hong Kong mercantilist?

Occam’s Razor suggests that the most plausible explanation is that high saving rates and a culture amenable to manufacturing in these regions leads to a large surplus in manufactured goods exports.

Wang doesn’t push his skepticism as far as I do, but he is rightly dismissive of claims that China got ahead by cheating, i.e., by stealing intellectual capital. Or that China is unable to innovate. He is critical of US attempts to slow China’s growth, correctly pointing out that these efforts are likely to backfire.

I wish Wang had applied that reasoning more broadly. This comment (on page 85) is a rare misstep:

So far, China hasn’t felt the economic pressure to abandon low-end manufacturing . . . That trend might not hold given escalating tariffs. But if Xi is successful, it means that other developing countries . . . will be unable to climb the industrial ladder that China reigns over.

Just as the US cannot hold China down, China is not the reason why areas such as Africa and South Asia have struggled to industrialize. In an article entitled “Morocco is now a trade and manufacturing powerhouse” the Economist suggests that even Morocco was able to become a major player in the global auto industry:

Last year Morocco became the biggest exporter of cars and parts to Europe, surpassing China and Japan.

The global economy is not a zero-sum game. Developing countries just need to get their act together. Smaller East Asia countries continue to do well despite China taking many of their former export markets.

I worry that Wang overlooks one big difference between manufacturing and housing/infrastructure. If America is to have good housing and infrastructure, we must build lots of housing and infrastructure. But that’s far less true of manufacturing, where many goods can be imported. We need to consume more housing and infrastructure, but we already consume plenty of manufactured goods.

To be clear, I believe that America does need a significant manufacturing sector, it would be foolish to rely solely on imports. And Wang is correct that we need to assure that we maintain enough manufacturing to have a sizable reserve of human capital with the skills required to scale up this sector if necessary (as we did in WWII.) The good news is that America is a manufacturing powerhouse, with a per capita value added in manufacturing that is higher than in Japan and more than double the level of China:

You say that graph doesn’t support my claim? Don’t forget that China has more than 4 times our population. And this doesn’t even account for the fact that a much higher proportion of Chinese workers are in manufacturing. Output per manufacturing worker is more than 4 times higher in the US than in China.

BTW, that graph appeared in an old Econlog post. When I went back and checked, it turns out that it was taken from an excellent Bloomberg article written by . . . Dan Wang and Ben Reinhardt.

And even in the specific area of arms manufacturing, we continue to do better than what one might expect from all of the handwringing by pessimistic pundits:

In per capita terms, places like South Korea and Italy do better at exporting weapons, but this graph excludes the huge domestic market for US manufactured arms.

In previous posts I’ve shown that the employment effects of trade are overrated, and that the loss of manufacturing jobs is mostly due to automation. China has already lost millions of manufacturing jobs in recent decades, and large parts of northern China can accurately be described as a “rustbelt.” Given that China’s population is likely to fall in half by 2100, and given that the share of workers in manufacturing will plunge due to automation and the rise of services (even without accounting for the effects of AI), China is likely to see its 100 million strong manufacturing workforce dwindle to just a few tens of millions later in this century.

To summarize, chapter 3 contains some refreshingly contrarian takes from Wang, but it is not quite contrarian enough for my taste. I disagree with the claim that China is a manufacturing powerhouse that the US needs to emulate, just as I rejected similar claims made about Japan in the 1980s (claims which have not aged well.)

Chapters 4-6:

This post is already running long, so I’ll only briefly discuss the second half of the book, with which I almost completely agree. Here Wang shows the dark side of an engineering society, and it is very dark indeed.

Most Americans have a negative view of China’s one-child policy, although I’ve met intellectuals who supported it. But I doubt that many people understand just how much suffering it caused. Wang shows how it directly or indirectly led to horrible side effects such as forced abortions, forced sterilization, infanticide, unequal sex ratios and child kidnapping.

Interestingly, the one-child policy did not apply to minorities, which comprise 9% of China’s total population. Of course it was first implemented in 1980, long before the global rise in right wing identity politics. And there have been recent reports of the Chinese government attempting to limit Uyghur births in Xinjiang.

It’s not even clear that the one-child policy achieved its primary goal, as China’s birth rate fell at roughly the same rate as did other East Asian countries. Indeed, China’s birth rate fell faster immediately before and after the one-child policy, than when the policy was in effect (from 1980 to 2015). Wang is also skeptical that Xi Jinping’s recent policy pivot toward encouraging women to give birth will have any effect. A recent article in Reason magazine reached a similar conclusion:

Spears and Geruso admit that they and other demographers have identified no policies that have ever lifted a country's total fertility rate once it has fallen below the replacement level. They point out that "population control has never controlled the population." To illustrate their point, they compare China's fertility trend under its one-child policy to the trends of peer nations. There is no discernable difference—fertility was falling at basically the same rate in each country.

Chapter 5 covers China’s zero Covid policy, which had some success at first, but failed miserably in 2022, after Covid mutated into the more contagious omicron variant. As with the one-child policy, China’s engineering approach to social issues led to mass suffering during lockdowns, albeit thankfully not as bad as the one-child policy. Here’s Wang:

Sometimes, the only thing scarier than China’s problems are Beijing’s solutions.

That is one of the defining characteristics of the engineering state. The Chinese government often resembles a crew of skilled firefighters who douse blazes they themselves ignited.

I suspect that many urban Chinese would regard zero-Covid as worse than the one-child policy, as the latter had its biggest impact in rural areas. Similarly, many urban Chinese view the Cultural Revolution as a bigger disaster than the Great Leap Forward, despite the fact that the GLF had a far greater death toll (mostly in rural areas). In China, urban residents are far more influential, and they determine what gets remembered as folk history. (Official history whitewashes almost everything.)

Chapter 6 looks at the cost of Xi Jinping’s move toward a more authoritarian regime that is reducing freedom of speech and travel (which admittedly were not all that free even prior to Xi taking office in 2012.)

Once again, I think it’s worth considering the urban/rural split. I suspect that most rural residents would see the Xi Jinping era as being freer than what came before, if only because they care more about the lifting of the one-child policy than they do about the ability to criticize the government on its Tibet or Hong Kong policies.

In my view, Xi Jinping’s decision to end term limits for China’s leader was his single biggest mistake, just as Putin’s decision to end term limits at roughly the same time was his biggest mistake. Let’s hope the US doesn’t follow their lead.

Final thoughts:

Dan Wang largely supports the current Washington consensus that the US needs to rebuild its manufacturing sector, especially in critical sectors such as chipmaking. But he also believes the US is going about it in all the wrong ways:

I spent years covering the twists and turns of these technology restrictions. The more it went on, the more I felt that the United States was committed to a strategy of destroying its scientific and industrial establishment—through prosecutions of scientists and cutting off the sales of chipmakers—in order to save it. Rather than realizing its own Sputnik moment, the United States triggered one in China.

China’s technology leaders have always bought American chips because they wanted to sell globally competitive products. They ignored Beijing’s beseeching to buy from domestic vendors for the simple reason that Chinese technologies were not good enough. But the Trump administration gave China’s tech leaders every reason to fear being cut off from American technologies. And so the US government fully aligned those Chinese firms that were previously reluctant to build up the domestic industrial base to Beijing’s self-sufficiency agenda. All the money and engineering talent that China’s most dynamic tech companies used to send to the United States were now staying at home.

Today’s FT confirms Wang’s fears:

China bans tech companies from buying Nvidia’s AI chips

In my view, the US is trying to do too many things at once. We claim that we are in a cold war with China for global manufacturing supremacy and ask the South Koreans to help us in that endeavor. But then our anti-immigrant agenda leads to hundreds of South Korean engineers being chained up like cattle and expelled over trivial visa violations (caused by our unwillingness to provide the visas that everyone knows are required for this sort of big project.)

There is much more of interest in this book. For instance, Wang is good at giving Western readers a sense of cultural differences within China:

And friends would tell me that Shenzhen, as in Silicon Valley, is a great place to found a start-up. A group of people would discuss an idea over dinner, divide up the tasks, and get to work the next morning. By contrast, in Beijing, dinner will feature interminable rounds of liquor shots, reckless bluffs about connections in high places, and uncertain follow-up afterwards.

Or:

Visitors from Shanghai liked to tease those of us in Beijing: “Why would you live in Pyongyang when you can live in Paris?”

There’s obviously hyperbole in these comparisons, but they make it easier to recall some very real differences.

One of the wonders of the modern internet is that it allows us to explore places that we’ll never visit. In years past, people would tell me, “Yes, coastal China is doing well, but the interior is still poor. That’s far less true today. Check out the video of Chongqing’s newest train station, the world’s largest.

There’s a cottage industry of this sort of video, and they do present an overly rosy picture—highlighting China’s best scenes. Some are probably paid for by the CCP. But there’s too much to easily dismiss. Consider this 27-minute drive along one of the world’s longest city streets (in Chengdu), or if that’s not enough a 57-minute video of the same street. They can be strangely hypnotic.

[If you’ve ever had spicy Sichuan food, it comes from the region around Chongqing and Chengdu, far in the interior.]

PS. A recent study found that China’s consumption is nowhere near as low as suggested by official figures, which implies that the share of GDP devoted to investment may be overstated. (A recent FT story reaches a similar conclusion.)