Globalization is not the Enemy: The Real Lesson From Trade with China
Among the Democratic candidates in the 2020 election, some, like Warren and Sanders, have carved out a position counter to the traditional consensus that free trade is unanimously good.
This has been fueled by the Democratic base who are skeptical of the benefits of globalization, and the Republican base’s feelings have been clear since the 2016 election. Trump constantly complains about unfair trade practices and has a penchant for levying tariffs to correct this perceived (though often imagined) harm to his supporters delight. For the first time, there is some agreement across the aisle that free trade is bad.
Standard Economics Asserted the Unanimous Benefits of Free Trade; Ignoring the Costs
This previous consensus on free trade among both parties was driven by the influence of economists that stressed the theory of competitive advantage, efficiency, frictionless adjustment, and technological displacement.
The central idea was that trade benefits countries overall because it allows them to focus on producing what they are relatively better at making (their comparative advantage), leading to increases in overall efficiency as countries specialize. Trade also provides access to a higher selection of goods at lower prices, expanding consumer surplus and increasing efficiency.
While some people may work in sectors that compete with these imports from foreign countries, they could easily move to other regions or industries that do not compete with foreign countries or move to industries that see demand expand due to globalization and trade (remember while countries might export specific goods to the US they often also import other goods they are relatively worse at making, again following the idea of competitive advantage). These import, export linkages from trade cascade through the economy, changing the allocation of labor across regions and industries.
This adjustment in labor markets would leave those directly affected by trade effectively unharmed on net because the adjustment is assumed to be frictionless (costless and quick), so everyone ultimately benefits from the lower prices and increased selection of goods.
Any residual harm that is leftover can be solved through redistributive taxes that transfer gains from those who benefit from trade to those who were harmed as many more benefit through lower prices and expanded choices than are directly harmed.
Furthermore, despite the theory of job losses from trade, most past research concluded that the real culprit of job loss was technology that automated away low skilled jobs in manufacturing, leaving only high skilled occupations (known as skill biased technological change). This was supported by the fact that the decline in share of employment in manufacturing in the US was fairly consistent since World War II, not correlating with increases in trade.
This view dominated the economics profession and spread throughout the political establishment to become almost gospel. Note though that this hinges on two crucial assumptions of (1) frictionless labor market adjustment and (2) adequate redistribution of gains, suggesting that poor theory and policy are to blame for the harmful effects of trade as described in more detail below.
The China Shock Changed Things, Showing Trade Had Larger Negative Impacts Than Expected
Autor, Dorn, and Hansen, three economists, received a lot of press for their paper titled the China Shock that exposed the deleterious effects that trade with China has had on specific parts of the US, illustrating that the above two assumptions do not hold in practice.
Frictionless labor market adjustment is not true for both structural and behavioral reasons.
In terms of structural factors, many workers are unable to relocate to different regions because of restrictive housing costs. Other structural factors like a lack of sufficient skills or credentials (potentially due to burdensome occupational licensing requirements) prevent workers in trade exposed sectors from switching to other industries. Retraining for different sectors itself is hard and often impossible because of a lack of access to affordable, quality education in the right areas.
On the behavioral side, many people feel emotionally tied to their local community because of friends and families, so purely on those grounds they may not want to relocate to other areas with more opportunity.
Both these groups of factors suggest adjustment to trade in labor markets is not frictionless, and indeed reveal the opposite: that it is quite costly for the individuals who are forced into that situation. Autor and coauthors find that in local labor markets that were particularly exposed to trade with China, they faced elevated levels of unemployment, depressed wages, and lower labor force participation for over a decade after the initial impact. This is far from the costless, immediate adjustment that traditional economic theory would have predicted.
Even among those who managed to move to a different region or work in a different sector saw their wages cut from previous levels. The reason is that many workers tended to join firms still exposed to trade with China, likely because these firms could make use of skills from their previous employment experience.
These shocks to individual sectors in a local economy were also amplified by supply chains and local markets, meaning the total effect wasn’t just from those losing jobs at companies competing with imports. Losses at a particular manufacturing firm in an area could harm other firms that relied on it as a purchaser of their product. Workers who lose their jobs are also unable to support the local economy through consumption, creating a negative multiplier effect.
We can see then that trade shocks, when large in magnitude from sources like the rise of China, can have a disproportionate impact on an area that relies heavily on a specific sector or individual manufacturing firm that is exposed to competition.
In theory, all this harm could have been ameliorated by proper redistribution. Autor and coauthors though find that in the past what support there has been was woefully inadequate to fully compensate for the losses due to trade.
Adding up all the different transfers (including direct aid for trade adjustment and other social welfare policies), the average person affected by trade would get $58, compared to expected losses of $549, so on net they were much worse off.
Despite This We Still Need to Promote Free Trade, but Also Couple it With Progressive Policies to Spread Gains More Uniformly
All this would seem to suggest that free trade really does harm American workers, and we should try to limit it where possible. This overview though misses the huge, diffuse benefits from trade in the form of lower prices and more access to goods (and indeed it also overlooks the peculiarities of the China Shock which saw the impact of a massive country’s rise that is unlikely to be replicated again).
The wrong lesson to take from the China Shock therefore is that trade is bad. It is not overall and indeed trade plus good policy could have heavily reduced the negative impact of China on local economies, leaving everyone better off.
The true lesson to take away from the China Shock is that local labor markets can be overly reliant on specific occupations, industries, or other concentrated factors, leaving them vulnerable to any disruptive shock (like the 2008 financial crisis for example). The disruption in the future is only likely to increase because of forces like technological innovation and climate change.
This means we need better federal and local level policies and institutions to accommodate the impact of these shocks, allowing local labor markets to withstand their impact and dynamically adjust over time. We need to make regional economies more resilient.
There is no silver bullet policy to accomplish this but rather a portfolio of policies that together can address the various sources of harm.
A straight forward redistributive policy to blunt the impact of job losses from trade or for any reason really would be to introduce a wage insurance program whereby if people lose their job and switch to another occupation with a lower salary, they will be compensated for the salary difference for a pre-specified amount of time. This would reduce the friction in finding a new job and especially ease the transition from well compensated manufacturing jobs to lower pay service sector jobs.
Another straightforward policy would be to expand access to affordable, high quality training programs. This doesn’t necessarily have to be college education (and there is evidence that is not always the answer) but rather more specific training like vocational school for in demand skills that are closely related to the skills workers already have to make better use of their previous experience.
Other policies that would reduce the friction in moving to new regions or industries would be reduced zoning restrictions in major US cities to allow for more affordable housing to be built and the removal of occupational licensing requirements to reduce barriers to entry into new careers.
In short, there are a plethora of policies we can enact in order to help communities adjust to shocks from free trade. These will in turn make these economies more vibrant and dynamic for future shocks, including recessions, that are inevitable and incredibly hard to predict.
We shouldn’t turn back the clock on free trade but rather expand it. It helps the US concentrate on its competitive advantage, benefit from lower prices and more consumer choice, and opens up an avenue to form tighter bonds around the world that can expand US foreign policy interests (like making trade pacts with Asian countries to counteract China). Combining free trade agreements with policies that address the damage it creates, using resources from the benefits, can ensure everyone is better off and in more stable communities.
