March 2023

March 2023
The China Shock

Leo Feler, Senior Economist, UCLA Anderson Forecast
David Dorn, UBS Professor of Globalization and Labor Markets, University of Zurich

March 2023 Economic Update

Economic Update with Leo Feler, Senior Economist, UCLA Anderson Forecast
  • Over the past few months, we’ve gotten some indications that the economy has grown faster and inflation has come down more slowly than many of us expected, which led us to believe that the Fed would tighten monetary policy by 50 bps at its March meeting.
  • But the past few weeks of banking failures have resulted in more macroeconomic uncertainty and a more cautious Fed. The Fed increased its benchmark rate by 25 bps to 4.75-5.0% on March 22 and changed its forward guidance to state that “some additional policy firming be appropriate” as opposed to its prior statement of “ongoing increases in the target range will be appropriate”. Some have called this a “dovish hike”.
  • As for the rationale behind this smaller hike, the FOMC stated that “recent developments are likely to result in tighter credit conditions for households and businesses and to weigh on economic activity, hiring, and inflation.” In other words, and as Fed Chair Jay Powell clarified during his press conference, recent developments in the banking sector have done some of the work of monetary tightening that the Fed would otherwise have done, and thus, the Fed doesn’t need to raise interest rates by quite as much.
  • Of course the issue here is that the monetary tightening from these “recent developments” is uncertain. More concern about the state of small and medium-sized banks might lead credit conditions to tighten further. If these concerns subside, credit conditions will likely loosen. This means the future path of the Fed’s interest rate policy is also highly uncertain. Right now, FOMC participants have signaled they may stop the tightening cycle after another 25 bps rate hike, but Chair Powell also said, “Participants don’t see rate cuts this year. They just don’t… Rate cuts are not in our base case scenario.”
  • Whether and by how much the Fed will continue increasing rates, and whether the Fed will truly hold rates steady for the remainder of the year, really depends on how the economy continues to evolve. Based on the Fed’s latest Summary of Economic Projections, the Fed has signaled that it is willing to let unemployment increase to 4.5% by the end of the year (from 3.6% currently), that it is wiling to have a few negative quarters of GDP this year (to get to a 0.4% YOY increase in GDP by the end of 2023), and that despite this economic slowdown and increase in unemployment, it will not lower rates. And all the while, it doesn’t see core inflation coming down to its target until after 2024.

A Conversation with Leo Feler and David Dorn


This month our podcast series features a conversation with Professor David Dorn regarding his research on the impacts of the China shock on US labor markets, voting behavior, and long-term impacts on mobility and migration.

Leo Feler: Welcome to the February 2023 edition of UCLA Anderson's Forecast Direct Podcast. I’m Leo Feler; I'm a senior economist at the UCLA Anderson Forecast. My guest this month is Professor David Dorn. David is the UBS Professor of Globalization and Labor Markets at the University of Zurich and he is currently a visiting professor at UC Berkeley. David, welcome to the program.

David, you have a series of papers on the economic and social impacts of the China shock. What was the China shock, and what was its immediate effect on US labor markets?

David Dorn: During the 1990s and the first decade of the new century, in the 2000s, there was a massive expansion of world trade in goods, and at the center of that expansion was the enormous transformation of China that went from being an essentially inward-looking economy that was not integrated into world trade towards becoming the world's leading exporter of manufactured goods. And as China had this dramatic transformation driven largely by internal reforms in China, this generated a rapid change in trade conditions for other countries such as the United States. Because this transformation and rapid increase in Chinese competition was unexpected, we tend to call it a shock. Hence the term “China shock.”

We’ve seen in just under two decades the amount of goods that the United States and other wealthy countries in Europe import from China increase dramatically, and as a consequence, many manufacturing industries in the U.S. have no longer been able to compete with the cheaper products from Chinese producers. That has had massive impacts on the U.S. labor market and on different geographic communities in the U.S. as well as on a social outcomes and political outcomes in the United States.

Leo Feler: How do you separate the effects of imports from China from the effects of automation and technology on the U.S. labor market? These seem to be happening at about the same time.

David Dorn: You’re right that these developments tend to occur around the same time. But interestingly, they did not all that strongly overlap in terms of geography. So, there were some places in the United States that were specialized, for instance, in furniture manufacturing, and where really the big impact was that furniture manufacturing, just like textile manufacturing, faced extreme competition from Chinese imports. Then you had other industries that were facing a lot of technological change, but oftentimes those were located in different places. So in separate work, I have measured the exposure to technological change by estimating the extent to which locations are specialized in so-called routine tasks and types of occupations and work that could potentially be automated. Interestingly, and also to our surprise, that exposure to automation and the exposure to Chinese trade was not all that strongly correlated across geographic areas. That allows us to estimate this effect of technological change and Chinese import competition separately from one another.

Leo Feler: What are some of the geographic areas that face the greatest import competition from China?

David Dorn: So an important thing to realize is that import competition from China did not affect all of U.S. manufacturing in an even way. Of course, when we think about the U.S. manufacturing decline, then maybe the first thing that will come to mind for many people is the decline of traditional manufacturing centers in the Midwest, such as in Detroit, for instance.

But Detroit is actually not a place that has a lot of exposure to Chinese competition, because the car industry that is characteristic for Detroit is not an industry that competes all that directly with Chinese imports. The same holds for industries like steel production, for instance, or industrial chemicals.

What does compete with Chinese imports is textile and apparel, furniture, the production of toys, also the production of home electronics. And the one place, for instance, where a lot of that came together would be the central region of North Carolina. Think of North Carolina's capital city of Raleigh. Raleigh had a sizable textile industry. They had an important furniture industry. They also had some electronics and manufacturing, and all of that faced extremely heavy competition from China.

Leo Feler: You also have another paper that looks at the impact of Chinese import competition on politics, focusing on the 2016 elections. What was the impact of import competition from China on the rise of populism and on this sense of discontent among blue collar workers that we saw in 2016?

David Dorn: Well, here I first have to explain why one might wonder whether these trade effects might impact political outcomes. I would think that these effects operate primarily through the impact of trade on the labor market.

What we found in our initial research some 10 years ago is that in those regions of the U.S. that had the kind of industries that faced a lot of Chinese import competition, we saw that the employment rates declined. We saw that there is a corresponding increase in the fraction of people who are unemployed or have left the labor force entirely. And we also saw that incomes in those places declined. These income declines were modest, compensated by additional government transfer payments. Once you're in a situation where you have lower employment rates and lower incomes, then a lot of other developments are triggered by that.

One research paper that I enjoyed a lot is a study that you did with Mine Senses on the provision of public goods in those regions that experienced more import competition from China. That brought the additional clever idea that once you have businesses struggling in those places, then also, the local communities will get less tax revenue partly because property values fall, and then there is less revenue from property taxes. That in turn means that these local communities might struggle to provide public services, like maybe public parks, good schools, etc. You showed also that there was an increase in criminality. And I think once you see this whole complex of outcomes—more poverty, more unemployment, more criminality, worse local services, or in additional work, even more mortality from drug related deaths—then it just becomes clear that even though manufacturing accounts for just a rather small fraction of U.S. employment today, it's less than 1 in 10 working age people working in that sector, it's still the case for the people in those communities that they will be negatively impacted. And so, the residents of locations that are heavily exposed to import competition from China, they understandably probably felt discontent with the way that things were going in their communities, and we have seen that voters in these communities have become more prone to elect people to Congress who come from either the far left or the far right of the political spectrum, with the effect of voting for the far right being the stronger one, and that includes that the residents of these communities have been more likely than residents of other communities that were not as impacted by import competition from China to vote for Donald Trump in the 2016 Presidential election.

Leo Feler: You mentioned in one of your more recent papers that the effect of the China shock plateaus around 2010, so we have a decade to study the longer-term effects of import competition from China. Just to clarify, what do you mean when you say that the effect of the China shock has plateaued, and then what are some of the more persistent effects that you see stemming from the China shock?

David Dorn: The China shock is really a huge transformation of international trade in goods. From the late 1980s into the early 2010s, the total volume of goods that are traded across national borders worldwide has roughly doubled. But then, in the early 2010s, we see that suddenly the expansion of world trade starts to slow down a lot. And indeed, the trade relationship between the United States and China no longer sees this massive increase in U.S. imports. There are two explanations for that. One is that this whole situation of a China shock clearly had some features of a “catch up” effect. So, once China moved towards a more capitalist system, an economic system that was more open to foreign trade, then the Chinese just had an enormous potential to catch up with the rest of the world.

It's important to realize that in the late-1940s and early-1950s, around the time when the Communists took over in Beijing, China was a relatively wealthy country by international comparison. It was just narrowly in the top third of richest countries in the world by GDP per capita. And then after a few decades of Communist rule, China had fallen from the top third to the bottom third of countries in terms of wealth per capita, and that means it started at that point in the 1980s, from a very low base and could easily then move up, partly because the Chinese also started to import a lot of foreign machinery, so they could improve their productivity tremendously by retrofitting their factories with much better, more modern machines. And so, then, what happens is at some point Chinese production methods start to look more like those in the West. They're also now using the same modern machines that the Americans or the Germans are using, and at this point, it becomes harder to still improve.

The second thing is also that around the early 2010s, the current leader of China, Xi, was taking over in Beijing, and he has clearly had an influence of slowing down the sort of market-oriented economic reforms, in some cases even reversing them. And that means that the whole sort of enormous export-oriented growth of China has slowed down a lot.

And now, even though this import competition is no longer growing a lot, one might think that those locations in the United States that have suffered during the 1990s and the first decade of the 2000s are now recuperating from sometimes adverse effects of these structural transformations. But to our surprise, we see that even nearly a decade after this China shock starts to wind down, these locations still have clearly elevated rates of a non-employment and lower wages. So it is the case that the scarring effects in the labor market are quite long lasting.

Leo Feler: You found something pretty interesting, which was a differential out-migration in the long run in response to some of the shocks that these areas in the US experienced. Can you talk about who moves in search of better labor market opportunities and who tends to stay behind in some of these areas that are affected by import competition from China?

David Dorn: Yes, you touch on a very interesting topic that really came as a surprise to many labor economists. I would say, some 15 years ago, the prevailing view among labor economists was that at least in the United States, workers are very geographically mobile. So there was a widespread view that if you lose your job in North Carolina then you just move to Boston or to San Francisco or to some other place where maybe there is still growth. And we showed that at least for this China trade shock, there just aren't very strong mobility responses of workers.

And even when we go 10 years past the shocks, the population responses are still quite muted. The groups, though, that are most likely to respond is, on the one hand, the younger people that might not yet be that involved in the local labor market or in owning a home. And also, you see more mobility of foreigners.

There seems to be some evidence that in the United States local shocks are sometimes mitigated by foreign immigrants that are not very tied to a particular location in the U.S. and are much more willing to move somewhere else. But one thing we found, interestingly, is that these adjustments by the foreign-born population has not been that effective in case of the China trade shock, simply because these locations affected by import competition were locations that didn’t have all that many foreigners to begin with. So, there is just not a ready stock of highly mobile people that could equilibrate these local outcomes by moving to areas with better labor market opportunities.

Leo Feler: David, thank you so much for joining us, and thank you also for this fascinating series of papers. One of the things that I really enjoy about reading your work is that you start with the China shock, and then you weave this well-researched narrative of what was the impact of the China shock on labor markets, then you rule out alternative hypotheses like automation, then you look at impacts on voting behavior, and finally you analyze long-term impacts like mobility responses and out-migration. It’s really fascinating to discuss this compilation of research you have worked on for over a decade.