China Never Stopped Managing its Trade

One standard criticism of Trump’s emphasis on bilateral trade—and reducing the bilateral trade deficit—is that it’s leading China back toward a world of managed trade.

Rufus Yerxa has noted (via Ana Swanson of the New York Times):

“It seems like those types of really simplistic purchasing commitment type of arrangements would actually reinforce state ownership rather than discourage it”.

Shawn Donnan of Bloomberg has made a similar argument.

They aren’t alone— this was almost a consensus view at the IMF’s Spring meetings.

The argument, though, in my view misses one critical point: China never stopped managing a large part of its trade.

Forget the formal structure of “trade”—tariffs, quotas and the like. They don’t matter much when the bulk of China’s imports are carried out by state owned companies, or by private companies that can only import with a license from China’s state. Go back and re-read Mark Wu's rightly celebrated article on China, Inc's challenge to the global trading rules.

When the state controls the firms that are doing the importing, a few phone calls can have a big impact.

That’s why China can shut down trade in canola with Canada without formally introducing any tariffs.

That’s why China can scale back its purchases of Australian coal without filing a “dumping” or “national security” tariffs case.

And that’s why—when the trade war with the United States started—U.S. exports in a number of goods simply went to zero (normally, a 25 percent tariff would reduce imports by more like 50 percent or something… the exact elasticity is debated).*

All the talk of China’s reform and opening up can be a bit deceptive in a sense. China did open up to imported components for re-export, and always has been willing to import the things it really needs and cannot produce at home. But decades after the start of “reform,” China’s state still controls the bulk of China’s imports of capital goods (through its ownership of the firms at the commanding heights of China’s economy, who make the key investments) and can influence the majority of China’s imports of commodities (through state trading companies like the state grain and oil seeds company, COFCO, and the state oil companies).

Consider the market structure of four of the biggest categories of U.S. goods exports to China: aircraft, soybeans, autos, and oil and gas. In three of the four, U.S. producers are already selling to China’s state, not to a private market.

Take aircraft.

Who buys Boeings?

China’s state owned airlines, and one financially troubled private airline (Hainan, whose parent now relies on state backing). And unless something institutionally has changed, they have to get the approval of the NDRC (China’s old state planning body) before any purchases are final.

China’s state is both Boeing’s single biggest market and one of its largest future competitors (through China’s ownership of COMAC, China’s national champion in civil aviation).   

And it throws its weight around: Rolls-Royce is looking to produce one of its jet engines in China to order to get key supply contracts. That's managed trade in my books, even if it is done without any formal tariffs or quotas on imported engines.

Take agriculture.

Who buys China's soybeans?

That is a bit more complicated. But the biggest buyer is almost certainly China’s old state grain trading monopoly, the China National Oilseeds, Grains, and Foodstuffs Company (COFCO). It no longer has a monopoly on China’s imports, but it controls an awful lot of the infrastructure needed to import soybeans, crush them into meal, and ship the processed soybeans to pig and poultry farms inside China. Sinograin—which manages the state’s strategic stockpiles—sometimes steps into the market as well. The smaller importers generally need an import license granted by the state.

Take oil and gas.

The big importers are still the big state oil and gas companies, which control the pipelines and own the bulk (but not all) of the refiners. It is a bit more complicated in LNG. As the Center for American Progress notes, China has allowed private firms to build the infrastructure for importing LNG (regasification terminals and the like). But the investors in major offshore projects still tend to be state firms. So while the “deal” on LNG with the United States looks to involve a state firm, I wouldn’t characterize that as a shift back toward state control. The state never really went away.

Take autos.

That’s the one sector where the state isn't the main buyer. It is also the sector where China’s imports have, at least to me, been a bit of surprise.

The formal barriers to trade are actually quite high. China’s tariffs on cars used to be 25 percent (20 percent for trucks). Even at 15 percent they are relatively high. 

And China—through the foreign JVs—has substantial domestic production capacity. It doesn’t “need” to import: all domestic demand could be met by domestic production.

Yet, the German luxury marks were able to build a decent business building cars in Germany and SUVs in the United States and selling them to China even with China’s high tariffs. I suspect production capacity originally built to meet U.S. demand for SUVs prior to the global crisis was repurposed after Chinese demand took off. And the German firms were able to do so in a sense because cars—setting aside the "made-in-China" Audis and VWs preferred by Chinese bureaucrats—are purchased privately.** All you had to do was pay the high tariff, not convince the state that they should import something that could be made in China…

Autos though are more the exception than the rule.

The reality is that China’s state influence over a wide range of large purchases, including large orders for a range of imports, never really went away, even after China joined the WTO.

That in some sense is the problem.

In sector after sector—particularly on the capital goods side—the ultimate buyer is the Chinese state.

Who buys telecommunications network equipment? The three state owned telecommunications companies.

Who buys railway equipment, like high speed trains? China’s railway ministry and its affiliated companies.

Who buys the electricity generated by large solar farms, wind turbines, and natural gas generators? The State Grid and the China Southern Power Grid.

Who buys the avionics and engines for China’s indigenous commercial aircraft? China’s national champion in civil aviation.

Who buys medical equipment? That’s a bit of a trick question, as there are some private hospitals, and a lot of the public hospital procurement is done by local governments and not central SOEs.

And, who sets critical prices, like the price of China’s currency?

That’s a bit of a rhetorical question, as I think China still manages its currency (to keep inside an undeclared band) even if its management isn't as large or as visible as it once was.

This all is a description of the world as it is, not the world as it should be.

But it gets to an important question.

In sectors where the state controls purchasing, and thus end demand, it isn’t enough to just reduce formal barriers at the border. It isn’t even enough to get rid of formal joint venture requirements. GE could set up its own independent gas turbine manufacturing firm in China, but it would still need to sell those turbines to power companies that are often state run and that ultimately sell to the state grid. Ask the various wind turbine manufactures who set up shop in China.

As I mentioned earlier, Rolls-Royce is considering setting up a wholly owned (I assume) subsidiary to make jet engines in China in order to get the contract to supply engines for the future widebodies of China, Inc.’s airlines. Airbus may agree to make the A330 as well as the A320 in China to try to increase its market share.

This is all part of what makes China a tough nut for imports to crack. China isn’t willing to give up the state’s control over the commanding heights of the economy in a trade negotiation. So if China doesn’t want to import something, it has a lot of levers it can use to limit import demand—even if there aren’t the kind of formal controls or taxes at the border.

The best critique of Trump’s deal—setting aside the argument that Australia and others have made that the United States is in effect selling out its allies to get a bigger share of a limited import pie***—isn’t that it returns China to a world of managed trade, where the state will have to reinsert itself in the economy.

Rather it is the criticism made by Kevin Book (and Derek Scissors): “If it can be negotiated by government fiat, it can be taken away by government fiat”.

On the other hand, it isn’t realistic to expect China to dismantle the core institutions of its mixed economic model in a trade negotiation. And until China’s state divests itself from the commanding heights of China’s economy and a lot of big investment decisions, well, China will retain the institutional structure to manage trade if it wants too.****

* Look at the size of the fall in China's imports of U.S. soybeans and U.S. oil and gas. The fall in auto imports was more modest, despite substantial tariffs. Soybean exports have slowly been recovering after the Buenos Aires deal, but remain below their "typical" post-harvest levels.

Oil and gas exports are also starting to recover.

And February auto exports were quite strong, all things considered.

** Audi became the preferred luxury brand of the Chinese communist party by making its luxury cars in China, and thus getting on the government’s procurement lists. There was a time when China's state may have accounted for one-twelfth of total global demand for certain Audis. After the anti-corruption campaign, China’s top bureaucrats had to trade down to VWs…

*** More U.S. LNG means less LNG from Australia, less pipeline gas from Russia or less Chinese domestic coal production; purchases of U.S. wheat and beef mean smaller purchases of Australia’s agricultural output. And Trump fanned these suspicions when he privately indicated he didn’t want to share the results of his pressure with America’s European and Pacific allies. That seems like a mistake, as there should have been scope for a combined push to raise China’s imports from everyone.

**** China’s WTO entry in effect created a dual economy. On one hand there were the—largely private—coastal supply chains that fed into China’s export machine. These firms helped produce goods for the global market. And initially those supply chains relied heavily on imported components, and generally used imported machine tools as well. On the other side, China’s state retained control over key parts of China’s domestic economy, as the (reformed) state sector remained the ultimate buyer of a lot of the capital goods that China’s economy needed to modernize. It used that purchasing power to encourage “localization” and the development of Chinese national champions in a range of industries. Fair enough, but that has had an impact on the scope for mutually beneficial trade. As China’s final demand has become more important to the global economy, so have the impact of China’s state firms’ purchasing decisions—and as China’s own productive capacities have increased, the scope to buy local has too. That’s my explanation for the tendency for China’s imports to fall relative to China’s GDP over time.