The China shock was real, and what free traders and tariff hawks each get right

Two statements are true at the same time. The first is that Americans bought a great deal of inexpensive goods from China, and that this left the typical family with more purchasing power. The second is that some American towns lost their factories to Chinese competition and did not recover. If you insist that only one of those is true, you are not doing economics. You are doing a campaign.

Conservatives split over this, and they split in good faith. One camp argues that open trade with China was a net gain and that tariffs are a tax on American families. The other argues that China took advantage of the system, hollowed out industrial communities and now threatens our security. I intend to give each camp the best version of its case, and then to argue for where I think the evidence points. The order is the usual one: define the terms, state both sides, test the claims, and reach a verdict.

The terms

Normal trade relations is the standard tariff treatment the United States gives to most trading partners. Before 2000, China's access to the American market under this status was reviewed by Congress every year, which created uncertainty. Permanent normal trade relations (PNTR) removed the annual review.

The World Trade Organization (WTO) is the body that administers the main rules of international trade. Membership commits a country to follow the rules and gives it the benefit of the other members' commitments.

The China shock is the label economists use for the surge of Chinese imports into the United States and other rich countries after about 1990, and especially after China's entry into the WTO, and its effects on local labour markets.

What happened

The key dates are these.

• In October 2000, President Bill Clinton signed the law that granted China PNTR, effective once China joined the WTO. Congress passed it, and the argument of its supporters was that trade would increase American exports, and would also bring China closer to open markets and the rule of law.
• On December 11, 2001, China formally entered the WTO.

After that date, Chinese exports to the United States rose steeply. The firms that did the exporting included many that competed directly with American manufacturers of furniture, textiles, toys, electronics and other goods.

Now the employment record. According to the Bureau of Labor Statistics, manufacturing employment in the United States stood at roughly 17 million in the year 2000 and had fallen to under 12 million by about 2010. Not all of that is China. Automation raised output per worker, a point that matters and that I will return to, and there was a deep recession. But the timing, and the particular industries hit, prompted economists to ask how much was due to Chinese competition.

The research

The most influential answer came from the economists David Autor, David Dorn and Gordon Hanson. Their paper, "The China Syndrome: Local Labor Market Effects of Import Competition in the United States," was published in the American Economic Review in 2013. The method was clever. They compared American local labour markets, such as particular counties and commuting zones, according to how exposed their industries were to Chinese imports, and tracked what happened to employment, wages and public assistance.

Their findings, in summary:

1. Local areas more exposed to Chinese import competition suffered larger losses of manufacturing jobs.
2. The jobs did not return quickly. Workers did not simply move to better jobs in other sectors or other places. Instead, unemployment, lower participation in the workforce and greater reliance on disability and other transfer programs followed in the hit areas.
3. Their estimate was that Chinese import competition accounted for roughly a quarter of the overall decline in American manufacturing employment between 1990 and 2007. Later work, including a 2016 study by Acemoglu, Autor, Dorn, Hanson and Price, put the net job loss from Chinese import competition between 1999 and 2011 at 2.0 to 2.4 million. The estimates differ with the method, and I would not stake a claim on the exact figure; read the papers for the details.

The significance for the debate is the second finding. The textbook model I described in Comparative advantage explained with Ricardo's wine and English cloth says that workers displaced by trade move to other industries. The evidence says that for many of them, in many places, they did not, at least not for a long time. That is a real correction to an easy optimism, and it came from economists who were not hostile to trade.

The case that tariff hawks make

The tariff hawk's argument, stated as its defenders would state it, has three parts.

1. The jobs were lost, and the towns were hurt. Statistics about averages hide communities. A plant that closes in a small town takes with it the suppliers, the diner, the tax base and the young people. The research confirms that the effect was concentrated and lasting.
2. China did not play fair. The charge here is more than low wages. It is subsidies to state-owned firms, the manipulation of currency in some periods, forced technology transfer, and the theft of intellectual property.
3. Dependence is a security risk. A country that relies on a strategic competitor for critical goods and components can be pressured, and could be cut off in a conflict.

On the second point there is an official record. In August 2017 President Trump directed the United States Trade Representative to investigate China's practices under Section 301 of the Trade Act of 1974. The USTR's report, issued in March 2018, found that China's acts, policies and practices related to technology transfer, intellectual property and innovation were unreasonable and discriminatory and burdened American commerce. The categories it reviewed included the use of ownership restrictions and joint-venture requirements to pressure foreign firms into handing over technology, discriminatory licensing terms, state-directed acquisition of American companies and assets, and cyber intrusions into American commercial networks. Tariffs under Section 301 began in July 2018. I describe the legal basis for that power in Who can raise tariffs? Article I, Section 232 and the delegated power.

I think the hawks are right about this point, and right that a country that accepts the trading rules should be held to them. Theft is theft, and a conservative who believes in property rights does not become indifferent to it because the thief is a trading partner.

The case that free traders make

Now the other side, stated just as fairly.

1. Consumers gained. Cheaper goods are not a trivial benefit. They raise the real income of everyone who buys them, which is everyone, and they help the poor most, because the poor spend a larger share of their income on goods. A tariff is a tax on those same families.
2. American firms gained from cheaper inputs and new markets. Many American manufacturers buy parts and materials from abroad. A tariff on those inputs raises their costs and hurts their ability to compete. And a larger world market means more demand for American exports in agriculture, aircraft, energy, software and services.
3. Automation, not China, is the main cause of lost factory jobs. Manufacturing output in the United States kept rising for years while employment fell, because each worker produced more. Blaming imports for the entire decline misreads the data, and protection would not bring the old employment levels back.
4. Tariffs invite retaliation and cost jobs elsewhere. Other countries answer a tariff with their own, and it is farmers and exporters who pay. Protection for one industry can mean losses for another.

The fourth point deserves a numbered answer from the hawks, and the third is the strongest, because it is correct. Automation is a major part of the story. The authors I cited do not say that China explains everything. They say it explains a significant share, and that is both consistent with the automation story and compatible with it.

Testing both claims against the evidence

Let me go point by point.

One: were consumers better off? Yes. The evidence that imports of manufactured goods lowered prices is strong, and the effect is not seriously disputed by the authors of the China shock research. The gain is spread across hundreds of millions of people and is hard to see, which is exactly why it gets less attention than a closed plant.

Two: were certain towns badly harmed? Yes. The research shows this. The gains from trade were not matched by compensation, and the adjustment assistance that exists is small and hard to use. That is a failure of policy and not of the theory. A nation that gains from trade has the means to help those it displaced, and it mostly did not.

Three: did China break the rules? On intellectual property and technology transfer, the official findings say so, and the pattern is documented. This is a matter of enforcement, and the proper remedy is enforcement: targeted measures against specific practices, in coalition with allies who face the same practices, and through the legal tools Congress has authorized.

Four: do broad tariffs fix it? Here the evidence is weaker for the hawks. A tariff on all goods taxes American buyers of everything, including the machinery that American factories need. It also does little to address the practices at issue, which are about technology and subsidies and not about the price of a shirt. A targeted measure against a specific abuse is a scalpel; a broad tariff is a hammer, and the hand that holds it pays for the swing.

A verdict

Here is where I come out, and I will say which parts are fact and which are judgment.

As a matter of fact, the China shock was real. It cost jobs, it hurt specific communities for a long time, and the benefits to consumers, though real, did not repair that harm. As a matter of fact, China engaged in practices that violated the spirit and in some cases the letter of its commitments, and the official record says so.

As a matter of judgment, I conclude that the 2000 decision was not a mistake in principle but a failure in execution. We opened our market on the strength of a promise that China would open its own and move toward the rule of law. When that did not happen, we kept the benefits and did not enforce the bargain. And we treated the people who bore the costs as an acceptable loss, which no conservative should accept. The family, the town and the church are where a nation's character lives, and economic policy that treats them as inventory is a policy that has lost the thread.

I also conclude that the remedy is not to swing to the other extreme. Broad tariffs as a permanent strategy raise prices for working families, invite retaliation, and rest on a belief that a Congress or an executive can pick the right industries to shelter. Conservatives know better than to trust that belief. The approach I favour has four parts:

1. Enforce the rules against specific abuses, with targeted measures backed by allies.
2. Protect a short, clearly defined list of goods that are essential to national defence, and say publicly why each is on it.
3. Keep the market open for the rest, so that American families keep the benefit of lower prices.
4. Invest in the places that bore the cost, through lower taxes, lighter regulation and the removal of barriers to new business, which I discussed in the regulation bill a ten-person company pays that a giant can shrug off.

The deeper history

It helps to put this in a longer frame. Hamilton argued in 1791 that a nation needs its own industrial base for its independence, and Smith had argued in 1776 that open trade enriches the nation. I read both closely in Hamilton's Report on Manufactures and the conservative case for tariffs. The China debate is the same argument with a different cast. The hawks are Hamilton's heirs; the free traders are Smith's. The wise conservative takes from each what the evidence supports.

There is also a lesson from China's own history, told in Xiaogang village signed a secret contract in 1978 and China began to grow. China's economic growth began when farmers were allowed to keep what they grew. The lesson that the Chinese Communist Party learned from that, and has partly unlearned since, is that markets work. It is ironic that the American debate over China now turns on whether we trust our own markets as much as they once did.

You can read more of the world picture on the Constitution page for the clauses that govern trade, and on the numbers page for official data you can check for yourself.

The case in brief

1. China received permanent normal trade relations from Congress in 2000 and joined the World Trade Organization on December 11, 2001, after which Chinese imports surged.
2. The research of Autor, Dorn and Hanson found that import competition from China caused significant, lasting manufacturing job losses in exposed local labour markets, which the textbook adjustment story did not predict.
3. The Section 301 findings of 2018 documented unreasonable practices on technology transfer and intellectual property, so the hawks are right that China did not play fair.
4. Consumers and exporters gained, and automation explains much of the loss of factory jobs, so the free traders are right that broad tariffs tax American families and miss the target.
5. The best conservative course is to enforce the rules against specific abuses, protect only a short list of goods vital to defence, keep the market open and help the communities that bore the cost.

What you can do: look up your own county's manufacturing employment on the Bureau of Labor Statistics website and ask your local officials what they are doing to attract new employers.