U.S. Strikes and Global Trends in Labor and Productivity

Tuesday, December 19, 2023

Professor of Practice, Harvard Law School

Professor Emeritus and Senior Fellow, Stanford University


Vice President for National Program and Outreach, Council on Foreign Relations

A. Michael Spence, distinguished visiting fellow at CFR, provides a global perspective on the changing landscape of labor and economic productivity. Sharon Block, professor of practice and executive director of the Center for Labor and a Just Economy at Harvard Law School, discusses this year’s strikes and the economic implications of increased collective labor activity in the United States. A question-and-answer session follows their opening remarks.


FASKIANOS: I’m Irina Faskianos, vice president for the National Program and Outreach here at CFR. 

We’re delighted to have over two hundred participants from forty-seven states and U.S. territories with us today. CFR is an independent and nonpartisan membership organization, think tank, and publisher focused on U.S. foreign policy. CFR is also the publisher of Foreign Affairs magazine. And, as always, CFR takes institutional positions on matters of policy. Through our State and Local Officials Initiative, CFR serves as a resource on international issues affecting the priorities and agendas of state and local governments by providing analysis on a wide range of policy topics. So, again, thank you all for joining us today. As a reminder, this webinar is on the record. The video and transcript will be posted on our website after the fact, at CFR.org. And we will circulate it to you as well.

We are pleased to have Sharon Block and Michael Spence with us to talk about U.S. Strikes and Global Trends in Labor and Productivity. I will give a few highlights from their bios.

Sharon Block is a professor of practice and executive director of the Center for Labor and a Just Economy at Harvard Law School. Recently, she served as a senior official in the Office of Information and Regulatory Affairs in the Biden administration. From 2017 to 2021, Professor Block led the Labor and Work Life Program at Harvard Law School, where she focused on labor law reforms to build a more equitable economy. Prior to that, she’s held various senior positions in government, including principal deputy assistant secretary for policy and senior counselor to the secretary of labor.

Michael Spence is a distinguished visiting fellow at the Council on Foreign Relations and clinical professor of economics at Bocconi University in Milan. He is also a senior fellow at Stanford University’s Hoover Institution and the author of the book, The Next Convergence: The Future of Economic Growth in a Multispeed World. And in 2001, Dr. Spence was a co-recipient of the Nobel Prize in Economic Sciences. So thank you both for being with us.

Sharon, I thought we could begin with you to give us a sense—give an overview of the increased collective labor activity in the U.S. that we’ve seen this year. If you could discuss the different strikes and the common threads, AI among them. 

BLOCK: Yeah, happy to. And thank you for having me. It is—I’m sure it will be a really interesting conversation.

So just set the stage, this summer into fall, I think, was a season like no other in recent years in the U.S. labor movement. There were approximately half a million workers who went out on strike in 2023. And a lot of that activity, again, was concentrated in the later part of the year. Another way of thinking about that is more than four million workdays were spent on strike instead of working. And to put it in context, that’s double the number of workers who went out on strike in the U.S. in 2022. So really a big upswing. But, to sort of pull back and put it in even sort of longer historical context, it’s a much, much lower number than what we saw decades ago in the sort of the high point of union density in the United States. You had millions and millions of workers out on strike, and a much greater part—share of the economy would be affected by those strikes. But in terms of, like, say, the last twenty to thirty years, this was a very significant year. 

The biggest strike in the United States this year was the SAG-AFTRA strike. That’s 150,000 union members were on strike. That’s a little bit of a funny number because SAG-AFTRA members are obviously a little unusual. They don’t go to work—typically they don’t go to work every day. But they are a very big group that withheld their labor, along with the Writers Guild. So you had Hollywood shut down for a significant period of time. The next biggest strike this year was the strike at Kaiser Health Care. Those are mostly SEIU members. That was about 75,000 workers who were out for three days. And then the strike that got certainly the most attention and was, I think, the third-biggest strike this year, was the United Auto Workers and their sort of novel strike strategy vis-à-vis the big three auto companies. Now, they did not take all of their members out at one time. So that was about 50,000 workers. There are more members in the UAW than that. But it was still a very significant number of workers, even with this sort of staggered strategy. 

So in addition to those three very large strikes we also saw strikes in the hospitality sector, in Las Vegas at the casinos, also L.A. hotels, and then in higher education. So the most of these strikes were really centered in the private sector. But we did have University of California, graduate student workers went out on strike. That was a very large strike. And then Rutgers University faculty and staff went out on strike. Now I would add to this an almost-strike, if you really want to think about how dramatic this activity was in the United States. The UPS workers—the Teamsters at UPS didn’t actually go out on strike, but took a strike vote, came very—like, within hours of going out on strike, at which time they were able to reach an agreement with the company. But it’s a similar dynamic of the threat of a strike that led to that agreement.

But say the theme among many of these strikes was that they existed—they happened between bargaining partners who have a very mature collective bargaining relationship. You think about the auto workers who have been unionized at the same three companies—you know, one of the companies has changed their name—but essentially organized at those three companies for almost a hundred years. These are not the kinds of bargaining relationships that have dominated sort of labor news over the past year or two, like Starbucks and Amazon, where you have new collective bargaining relationships. We didn’t see strikes among those workers. We saw them in these very established relationships. The other theme among these strikes, really almost universally, were very, very big wins for workers. They settled these strikes with agreements that, I think, were objectively viewed as very advantageous for workers. You saw very high levels of public support for the workers in almost all of these strikes. 

And then, to your point about AI, these are also strikes that happened, for the most part, in sectors that are in big transition. In some, because of the introduction of AI. That was obviously a very big theme, a big factor in the Hollywood strikes, but also other transitions. In the autoworkers strike you had the issue related to transition to an EV future played a big role, in healthcare that’s obviously an evolving field. So this idea of there being a big transition and workers using their power through their strike in order to get contracts that help them have more of a say in the future.

And then I would say one last theme that was very prevalent in many of the strikes was the sort of rhetorical and motivating theme of workers wanting to have their fair share. You heard that phrase come up a lot. So we’re talking about sectors where the companies had had a recent history of very high profits, workers who were locked into collective bargaining agreements that they had negotiated sort of before the pandemic. So if you think about, like, UPS had very, very high profits during the pandemic. The Teamsters were working under a contract that didn’t anticipate that level of profits. You have—the auto companies were also coming off a couple of years of very high profits. And so you have this theme of workers really wanting to get their fair share of this increased revenue and profits that they saw coming into their—into their companies.

The last thing I will say is just if you want to understand just sort of how positive this strike season was for workers, you just have to look at the UAW contracts. I mean, there are so many things about this strike that were just groundbreaking, or at least groundbreaking as of the past few decades. You saw wage increases of 25 percent for permanent workers and 150 percent for some of the temporary workers. You had really novel provisions in the collective bargaining agreement that they eventually signed to keep open or reopen auto plants. We’ve never really seen that before in a collective bargaining agreement. And workers preserving the right to strike over any other plant closures. As I said, you got this foothold in the EV future in agreements for the companies to recognize the union in these EV battery plants.

And so, it was just a really remarkably positive contract that ended the strike in the auto sector, really transforming the UAW to be able to say, again, that a job in the auto sector equals a good middle-class job. And we’re seeing now the autoworkers taking that message to the nonunionized companies—Tesla and the transplant companies—to say, look what we got for workers at the big three. Wouldn’t you like to have this too? And you’re seeing actually these companies already responding by raising wages. So it’s also a strike that has had pretty significant ripple effects already. One thing to watch in 2024 is how far those ripple effects go, how successful they are. Will this season of successful strikes for workers actually lead other workers to want to organize a union in their own companies, in their own sectors, maybe even beyond the auto sector?

And, again, we did have some groundbreaking provisions that came out of these strikes around AI. The Writers Guild, most significantly. You had agreements that AI can’t be used to undermine the writer’s credit, requirements for studios to disclose if they’re giving any material to writers that was generated by AI. But then also, in a sort of more positive embrace of AI, the right for writers to choose to use AI as a tool as part of an agreement with the studios. SAG-AFTRA, the actors also got provisions sort of protecting their images from AI replication without their consent. And the Las Vegas—the hospitality unions also got provisions guaranteeing them advanced warning of any new technology rollouts that were going to impact jobs and training for jobs that are altered by AI. And, really importantly, protections from certain types of AI that enables surveillance within the workplace, something that was very important to hotel workers who have been increasingly surveilled in their work.

So there is a lot to dig into. I’m going to stop talking so we can get to some questions, because there’s really—could go on and on because it was such a fascinating period of time.

FASKIANOS: Fantastic. Thank you so much, Sharon. 

Michael, let’s go to you to pull out a little bit and talk about the global trends you’re seeing, and the implications for the future workforce and labor movements. And you just recently authored an article in our magazine, Foreign Affairs, The Coming AI Economic Revolution, with James Manyika. So perhaps you could talk a little bit too about the AI piece of this as well.

SPENCE: Well, thank you very much. And I’m, you know, like Sharon, very pleased to be with you.

So let me approach these things, you know, at a sort of slightly different level. That three-decade period that Sharon referred to is a period in which a massive amount of productive capacity was introduced into the global economy, largely as a result of emerging economy growth. And that had one very large negative effect, which was it, you know, created options for, you know, labor arbitrage and decreased the power of American labor. So unions declined, you know, the middle class got hollowed out to some extent, and so on. That force is fading. It’s not over, but it’s fading. There’s lots of evidence of that. You know, for at least two decades, probably more than that, we lost employment in the manufacturing sector. That stopped in the last decade.

And then—but then there’s some other trends that, you know, kind of reinforce this. So when I look, you know, I see aging populations. Seventy-five percent of the global GDP is produced in in countries that are aging rapidly. You know, the great financial crisis caused some of our older fellow citizens, like me—not to retire. Now they’re retiring in droves. You know, when I look at the American economy all—most of the big labor, you know, employment sectors have labor shortages, right? I mean, it’s clear that on the underlying economic fundamentals, labor’s power position vis-à-vis their employers has increased dramatically. Some of this shows up in unionization. Some of it just shows up and in bidding for, you know, talent in a way that basically companies didn’t have to before—or, employers in general.

So I think this is basically a good development. I expect to see, you know, several attractive trends. A reversal, maybe not a dramatic one, in the trends in inequality on the income side, which would be very good thing because it had gotten pretty extreme over this three-decade period. You know, I think we will see productivity increases because when you’re short of labor it’s sort of natural to start looking—the incentives are much stronger to look for productivity-enhancing things. And if that’s done in a way that makes—you know, puts management and labor in a collaborative position, seeking for ways that are mutually beneficial to do it, that’s also a good thing.

On the negative side, you know, this is—you know, for the first time, really, we live in a supply-constrained world. I just—you know, at the risk of telling people what they already know, after the great financial crisis we’ve had—and for a longer period than that—we’ve had essentially no sign of inflation whatsoever. And we had no sign of inflation, in spite of zero interest rates and massive infusions of liquidity into the economy to try to precipitate a recovery after the balance sheet damage that the great financial crisis caused. And as a result of that, people have kind of gotten used to the notion that, you know, the cost of capital isn’t very high. So for people who are operating in state and municipal governments, I think, you know, there’s—nobody knows for sure. And we have a big inflation fight on, led by the central banks. Not just in the United States but in the U.K., and in Europe, and in other places. China being a fairly dramatic exception to this.

We’re likely, in my view, to emerge from this with higher real interest rates. I don’t have any doubt that the central banks will get the inflation eventually under control, because they’re determined to do it and their credibility depends on it. That’s their job. But when we come out, I think we’re going to have, you know, lower sustainable debt levels, higher cost of capital, lower multiples, lower valuations for many assets. This will have mixed effects. You know, the cost of funding, certain longer-term investments is going to be a little bit higher than it was before, maybe even more than a little bit higher. On the other hand, from a distributional point of view, you know, when—in the period—the decade after the great financial crisis, the one thing that just ballooned in value was the assets. And that favored people who, you know, own a lot of assets. So it didn’t do wonders for the distributional features.

So I think on the whole, if you sort of look at—I mean, there’s a lot—a lot of other factors, you know, that are affecting this. The global supply chains are, you know, collapsing—or, being fragmented. We have a major strategic competition, you know, with China underway. Economic policy, from an international point of view, has tipped toward, you know, various kinds of security—national security prominently, but also economic security, here in Europe energy security, food security, and so on. And this is causing, you know, policy to reinforce a trend in the global economy that’s very visible now, which is diversification in pursuit of resilience. 
And the policy is reinforcing it and saying: We have to do some of this at home in a way that we didn’t pay attention to before. We lived for three decades, those three decades, in which the way global—the global economy was constructed was basically on the basis of economic efficiency and comparative advantage. And that’s no longer true. So we have homeshoring, friendshoring, nearshoring, et cetera. All of which are transforming the structure of the global economy. And for the most part, I think, in ways that favor, you know, domestic—our domestic fellow citizens, and especially labor.

Briefly on AI. So, we’ve had a sequence of breakthroughs in AI that go back, you know, a decade or a bit more. Language recognition. You know, image recognition was a stunning set of breakthroughs that, you know, occurred roughly around 2015-2016. But the one that’s really gotten people’s attention is generative AI, the large language models and the like. So there’s several things to say about this. And I’ll try to be brief. One, we’re not at the end of this. These folks aren’t finished. So what’s coming next we don’t know. I suspect that we will see significant advances in robotics as a result of the fact that gen AI allows you to basically talk to machines in a way that they understand. 

The gen AI is distinctive in the sense—in two respects that I think are important. One, unlike any other previous version of AI, they switch domains easily. By that, I mean, you know, you can talk to it about the Italian Renaissance and then switch to math and then it’ll do computer coding, you know, and whatever, right? Now, there’s lots of quirks, you know. These systems so far have hallucinations. They make stuff up. And I mention that for a reason. You know, it’s not—when you look at it carefully, it’s not sensible to think that these things should be fully, you know, allowed to operate on their own, right? They’re just not that flawless. You know, there’s a famous story in America, you know, a lawyer, slightly incautiously, prepared a legal brief entirely using ChatGPT, and handed it in. Well, ChatGPT made up all the legal precedents. And this gentleman is, I think, in some serious trouble as a result with the courts. 

So the way I think about it, and I’m not alone in this. I mean, James and I wrote that paper. We think that the right model is powerful digital assistant or machine-human collaboration, right? And you have to work that out. But let me say, you know, right at the top, there’s just overwhelming evidence that whenever you mention, you know, AI, people think, automation. They think they’re coming for our jobs. A hospital administrator stands up and starts talking about AI—and, by the way, AI is going to be transformative in biomedical and life sciences, which is not our topic for today. But it’s just one of the many places where the footprint over time will be felt. We have to overcome this bias. So the implementation matters. 

You know, unions representing people and having a voice in which they participate in conversations about what the AI is supposed to be doing and how it will change the jobs, and which parts are acceptable or not. But I think in the course of it we can sort of get rid of this—what I call the automation bias. Erik Brynjolfsson at Stanford calls this the Turing trap. Alan Turing proposed that we evaluate our progress in AI by asking the question: Can we produce a machine that when a human being interacts with that machine, not looking at it but talking with it, it thinks it’s interacting with another human being? And so we haven’t got there yet, but we’re working on it.

Second, one small step. Almost all AIs are benchmarked against human performance. So when they declare victory in image recognition, it’s when it passes the average human, and so on. It’s the next small step that’s dangerous, which is, you know, well, once the machine passes the human, why don’t we replace the human, right? That’s where the AI—the automation bias comes from. And it’s just a mistake. Now, there may be a time in the future when these machines are so good that automation is a more serious consideration. But right now, they’re powerful digital assistants. They can sometimes do things that humans can’t do. Sometimes they do them, you know, in a way that’s just on par. But they—but I think the promise here is if we do this right that we’ll have the potential—not next year, not the year after, but maybe by the end of the decade—we’ll start to see, you know, impacts of this and in terms of productivity that are actually, you know, enhancements to the way people work and how they view their employment.

To get there, we got to get rid of the automation bias, which is very deep. And we need one other thing. We need access. So right now, we’re in a period of intense exploration and experimentation. Who’s doing this, right? The answer is the companies with the resources to do it, you know. But if we’re going to have this broadly beneficial in society, available to small businesses, to local governments, and so on, it has to diffuse widely and well beyond, you know, the kind of entities that have the capacity to invest tens of millions of dollars in it. There’s a role for government in this. 

And I want to conclude with this, because we’re talking to, you know, important government officials in our economy. There really is a role, you know, in ensuring diffusion and broad-based access to these tools once we’ve decided, you know, in rough and ready terms, you know, how we’re going to try to use them. It’s really important, both for the purpose of getting the productivity surge but also, you know, for preventing—you know, in past—there’s studies of this at the McKinsey Global Institute. In past, you know, episodes of digital, you know, transformation, they’ve studied adoption. So and what you see is a pattern of divergence. So the tech sectors way is way up top, and finance is not far behind, and then you drop down and find sectors that, you know, are lagging seriously in this respect. This is the pattern that we do not want to repeat on this round. 

So I think there’s huge potential. There’s some downside risks. James and I would say that, you know, it’s important to pay attention to the misuse of these things and the downside risks, just as there is with any powerful technology. I mean, gene editing is terrifying if used in the wrong way, just as AI is as well. But there’s the positive agenda as well.

FASKIANOS: Thank you both very much. And now we’re going to go to all of you for your questions.
We’ve got the first written question from Riley Nye: Hasn’t automation already replaced tons of manufacturing jobs?

SPENCE: Sharon, do you want to take that, or?

BLOCK: I mean, certainly there has been displacement. Earlier this year I visited a Ford auto plant. And it is very, very—a very, very different place than certainly historical documentation of what the manufacturing process looked like. There are a lot fewer workers. You see that also reflected in—just to stay with the auto sector—in the number of UAW members in that sector. The UAW has actually diversified their membership a great deal. They do a lot—as many people on the call may know if you’re involved with higher education—they represent now many, many graduate students and other employees of higher education institutions. So, yes, that has happened. But the displacement conversation is obviously not over. And there is, I think, concern about additional displacement of workers if robotics and those kinds of productivity enhancements, or whatever the right euphemism is, continue.

But I do think in the shorter term, the bigger concern actually should be for workers is the way that automation is being used in workplaces to enhance not just productivity but also employer domination of workers. These surveillance issues you’re seeing, especially if you follow, like, the logistics sector. The intensification of work that is enabled by the kind of productivity tracking that AI has enabled. I think these are the changes in the workplace that people are feeling already, even before we get to this question of whether AI is ever going to—or when it’s going to be good enough to replace more workers.

And so that’s a place where I think the regulatory attention really needs to be paid. And if you look at what the EU just did—we don’t know the details yet of the AI Act that the EU Parliament just passed—but there is a lot of attention to those issues. And, in fact, the workplace is designated in that legislation is a high-risk forum for the introduction of AI, not because of the displacement issues but because of the intrusion into sort of personal privacy spheres for working people, and this potential for new safety and health issues to arise from a misuse of AI in the workplace.

FASKIANOS: Thank you.

I’m going to go next to Justin Freeman, who is the director of community affairs for New York State Assembly: Could you share more about how today’s strike actions compare to before, say, thirty, forty years ago? You mentioned four million strike days.

BLOCK: Yeah. So, again, we haven’t had a year with four million strike—it’s actually more than four million at this point. That doesn’t capture the full range of the UAW strike days. But I just couldn’t find a more recent calculation. That’s a little bit of a hard calculation to do. We haven’t seen a four million strike day year in a long time. So, say, ten to fifteen years ago. But if you look at, like, in the 1930s, so the period as we were legislating the right to join unions, you would have—there were years in the 1930s, early 1940s, where you had thousands of strikes in the United States. 

And, you know, it’s hard to compare numbers because our economy is obviously so much bigger now. Our workforce is so much bigger. But if you could imagine thousands of strikes. It was really a completely different scale. I mean, if I could show you a graph from, like, the ’40s to today, you would see a line that just really dramatically falls off as we entered this period, you know, that we’ve been talking about, like the past thirty years of a real decrease in the strength of the labor movement. You saw a commensurate decrease in the number of strikes.

FASKIANOS: OK, thank you.

Shawn has asked a question. China was mentioned. How big of a threat do you feel China is, with their housing, population, and debt crisis?

SPENCE: OK. So I think everybody knows that China, you know, has had kind of a pretty impressive forty-year run. It was one of the poorest countries in the world in 1980. It has exhibited growth rates of, you know, 7, 8, 9 percent on a sustained basis that, you know, causes, you know, the size of the economy and incomes to double faster than every decade. That’s not something an advanced country can do. I think China is now in a very difficult sort of position in terms of transformation. And the economy is in trouble. 

So they have major, major excess capacity in real estate, and a lot of non-performing loans, and whatnot. This directly affects the Chinese economy because while this is, to some extent, true in other places, the household balance sheet—meaning wealth—is heavily dependent on real estate, right? Once they started buying houses and so on, they just owned more real estate and less kind of other kinds of assets than almost anywhere else in the world. And so when the real estate values decline, or there’s some uncertainty, or it gets shaky, or, you know, the apartment that you bought in advance doesn’t get built, it causes a major shakeup in confidence. 
The fiscal system needs major reform in China. The municipal governments are essentially flat-broke. They do not have the normal sources of revenue that our municipal governments do. And they are responsible for delivering services that are in excess of their capacity to finance it. There’s no more land. They used to do it by selling land. There’s no more land to sell, or not enough to finance themselves. But I think the really sort of serious challenge, in addition, in China is that the pattern of off again/on again and fairly aggressive regulation, which you can see in the tech sector but it’s broader than that, has caused a loss of confidence among investors. 

And by that, I don’t mean foreign investors. I mean, everybody, including the domestic investors. And so with the household spending, you know, a little bit on hold, and the private sector investment, you know, kind of on hold because they’re not sure what their place in the sun is, that, you know, there’s a significant slowdown. The numbers in China will look OK because the previous year was a disaster. So when you’d show up growth numbers, you know, when they were in zero-COVID, it was, you know, unimpressive. So the numbers look better than the actual situation.

Having said that, you know, it is in many ways—you know, in human capital, in science and technology, and whatnot—they’ve made huge investments in it. So I don’t want to leave the impression that this is a kind of, you know, permanent disaster at all. They can pull this out. And it’s an economy—it’s a very large economy, the second-largest one in the world. And it will be, if they right the ship on these—what I think of as short- and medium-run challenges—it’s a, depending on how you think of it, a powerhouse and potential major competitor.

FASKIANOS: And, Michael, just to follow up on that, a question from Alan Schneider, who is legislative director in the Office of Maryland Delegate Chao Wu: You know, given the relationship changes between the U.S. and China, how are the changes affecting wage, A1, and inflation here in the United States?
SPENCE: Very good question. So in terms of, you know, the—we are in—our national security, you know, driven policies are bringing more stuff home. And in addition, China is now an economy with a per capita income of $13,000. You’re not going to make, you know, the cheapest labor-intensive, process-oriented manufacturing and assembly stuff in China for very much longer. There is no real substitute for China. There are other countries. And some of them are benefiting—Vietnam, Bangladesh, you know, Mexico has a major opportunity as China sort of, A, gets into kind of conflict with us and, B, you know, we move both businesses and governments in behind with policy to move stuff away. I think on the whole it’s slightly inflationary. But it’s good for, you know, labor—meaning, our labor.

FASKIANOS: Terrific.

Let’s go to—sorry. Going to sell in Selin Zorer: What proactive steps should federal and state legislators take to ensure that the emergence of AI benefits the public?

SPENCE: Sharon, do you want—do you want—

BLOCK: I can—I can start. I mean, you know, I think one way to ensure that it benefits public—I mean, most of the public has to go to work every day. And so thinking about the ways to protect workers from some of these abuses and excesses is really important. The other—I think the other area where I’m really interested is while we see some paralysis at the federal level in terms of legislating around the introduction of AI into the workplace, I think there is much more of an opportunity for state and local governments to step in. California is obviously very engaged in their legislature in thinking about guardrails for AI in the workplace and in other domains. 

But it’s really important that as this regulatory forum moves to state and local levels, that there is an attention to making sure that the people who are going to be most affected, that working people have an opportunity to have a voice in how this regulation develops. And so whether that’s bringing in the labor movement, finding other ways to ensure that working people are participating in these really important conversations, I think is going to be critical. And I hope we’ll see sort of interesting and innovative approaches as more states feel compelled to get into the game. Because we are probably not going to see, you know, significant federal regulation or legislation in this space.

FASKIANOS: Thank you. Next—

SPENCE: Irina, could I—

FASKIANOS: Go ahead, Michael, absolutely.

SPENCE: This is—I don’t want to repeat myself but, you know, there’s a positive agenda. You know, a lot of the, you know, management that affects people’s lives is done at the state and local level. Not, you know, the kind of stratosphere where some parts of the federal government operate. And, you know, I think, you know, thinking carefully about where we’re going with these technologies and how you help people, you know, become comfortable with them, productive with them, and so on, is a hugely important part of the agenda. And I can’t think of more important entities than the state and local governments, you know, the community colleges.

You know, the education system as a whole seems to me to be, you know, where the conversation needs—you know, a fair amount of the conversation needs to occur. So, again, I don’t want to, you know, minimize the importance of preventing downside risks and misuse and so on. But I think walking into the world, you know, without a coherent set of programs to help people—you know, if we are going to have these transformations in one form or another. That it’s way too powerful, these technologies. So I think the challenge is to do it right, rather than resist them.

FASKIANOS: Fantastic.

I’m going to go next to Justin Freeman, director of community affairs at the New York State Assembly: Is there any correlation between interest rates and strike actions? Can strikes be anticipated through economic indicators?

SPENCE: Go ahead.

FASKIANOS: You can take it. Who wants to start?

BLOCK: I think Michael mentioned the most important factor, which is a tight labor market. I mean, that is clearly connected to this upsurge in labor activity, both in the strike activity and then also in this renewed commitment to organizing. There’s just—there’s a lot of risk under our legal system. There’s a lot of risk to workers who try to organize a union, who go out on strike. We have a law that’s really deficient in terms of protecting workers who engage in that kind of labor activity. And so a tight labor market gives workers the confidence that if they are retaliated against for taking this kind of activity, that they can find other jobs. And it’s really as simple as that as to why you see that correlation between a tight labor market and increased union activity.

So I think that’s the most important factor. I think the issue of interest rates is just whether the Fed was going to raise interest rates enough to start driving that unemployment rate up and creating slack in the labor market, which then would have taken some of this dynamic—diluted this dynamic so that workers didn’t have that same confidence in their ability to find other jobs when they take the risk of organizing or striking.


SPENCE: Yeah, I mean, essentially the same. I mean, so we—you know, the supply side of our economy, and the global economy has just changed dramatically, right? So it used to be almost infinitely elastic. You could have a surge in demand and, you know, somewhere somebody produced enough to meet it. That’s just not true anymore. That’s why we have labor shortages, as Sharon says. That’s why labor power is increasing. And as for inflation, you know, the trigger, you know, as we came out of the pandemic with a predictable surge in demand, and the supply side constrained by, you know, aging—you know, all the things we talked about. You know, we had a demand and supply imbalance. It was the trigger for inflation.

Now, inflation can develop a life of its own, you know, once it goes on for long enough. But, you know, the economists look at this and say: When have we seen interest rates go up this fast and this high and not seen, you know, labor market problem? We just—you can go back a long way and try to find out an example of this. So, you know, this—what this tells you is that, you know, we have fundamental structural changes underway in the economy. And they—and the relationship between the labor markets and the inflation is that, you know, it triggered the inflation because the supply side couldn’t keep up.

Now, what’s going on now, and I’ll just end with this, is, you know, the central banks, you know, can’t operate on the supply side of the economy. So they’re basically raising interest rates largely to reduce aggregate demand and get rid of that imbalance. And so far, they managed to do it without, you know, producing unemployment increases of any significant magnitude because there are labor shortages—short version. 

FASKIANOS: Great. Thank you.

I’m going to go next to Aaron Tebrinke, who’s legislative assistant to Leader Koehler of the Illinois Senate: After 148-day strike, Hollywood screenwriters secured significant guardrails against the use of AI in one of the first major labor battles over generative AI in the workplace. A battle for automation against AI, automation was won by labor. But what protections will workers have to keep up with AI tools in the marketplace that are not regulated for privacy?

SPENCE: Sorry, can I—the concern of the writers was that, you know, they were going to get displaced, you know, by the use of, you know, the kind of generative AI, the large language models. That was, like, I don’t—that’s a bit of automation, but the underlying concern was copyright, right? Which is a major issue, right? Because it—you know, gen AI is trained on the entire internet. They just go read everything, you know, at speeds that exceed human capacities. So the question is, well, what’s the relationship between that and all the imaginative content that these and other people have produced that the AIs just hoover up? So that strike had multiple dimensions to it, and not all of them had to do with automation, for sure.

FASKIANOS: Sharon, anything to add?

BLOCK: Yeah, I think that that’s right. Obviously, we’re seeing litigation by content creators, many of whom are members of the Writers Guild, in order to get at this issue of their intellectual property rights vis-à-vis the use of these—the use of that content by the large language models. So I think we are going to continue to see many different fronts in the introduction of AI into the workplace, and as it impacts workers in different ways. So but to just to answer the, the part of the question about privacy, we have a very, very weak privacy regime vis-à-vis the workplace in the United States. And so you really—in the private sector. Now, that’s different in the public sector because you have a constitutional dimension to privacy in the workplace with public sector employers. So some of this might sound—might sound different than your own—than your experience, since folks on this call are from the public sector. 

But in the private sector, we don’t really have an institution of privacy protections—as we now have AI surveillance of things like, you know, your email. There are employers now who very easily can just scrape every email that you write to find out all kinds of things about you, and you probably don’t even know it. That can watch you through the camera on your laptop when you’re working from your home. So I think these privacy concerns aren’t new in the workplace. But I think they’re going to be appreciated by, I hope, policymakers, but also by workers in a new way, as we see different uses of AI in the workplace.

FASKIANOS: Great, thank you.

I’m going to go next to Nate Belcher, who is a fiscal analyst for Arizona Joint Legislative Budget Committee: UAW included a thirty-two-hour workweek with no pay reduction as one of their bargaining points in their recent negotiations. Do you think that reducing the length of the workweek will become a more popular demand from labor in the coming years?

BLOCK: Yes. I think—I mean, I think we’re seeing it already. I mean, I would say just a few years ago there was almost no serious discussion of a four-day workweek. That is now an issue that is on the table. I don’t know of any workers who have secured a four-day workweek through collective bargaining. There are certainly employers who, of their own volition, are experimenting with shorter workweeks, sometimes with four-day workweeks. You know, I don’t think that many people thought that the UAW contract at the end of the day would actually include a thirty-two-hour workweek. 

I think it was put on the table as just another way to discuss hours. I mean, what was really an issue in the UAW strike I think around hours was the fact that many, many workers were being forced to work a lot of overtime. And even if they were getting paid for that overtime, it was having such an impact on their quality of life that it was really an entree to talk about what it is like to have those kinds of time demands, and what workers want in terms of having some kind of balance in their lives to be able to do with their time what they want. But I think the thirty-two-hour workweek is a conversation we’re going to continue to see bubbling up.

FASKIANOS: Thank you. 

Next question from Paul Egnatuk, who is the legislative aide in the office of the Michigan State Representative Jim Haadsma: I’ve heard recently of brick-and-mortar type investments stalled because investors are enamored with AI ventures. Can you recommend sources of research on the impact of private capital going toward AI development and/or where capital may be short for other pressing needs?

SPENCE: Right, this is complicated. I mean, so, you know, there’s a massive amount of money going into AI. So some of the valuations are probably a little bit off the chart and, you know, that’ll get corrected over time. Some of us will remember the internet bubble, which had some of the similar characteristics. But that doesn’t mean there’s nothing there. But, you know, if you look at, I mean, vis-à-vis the previous subject, you know, hybrid working is becoming a very prominent feature of a subset of the economy where you can do that, right? And, you know, if you go into New York now and go into an office on Friday—you know, you’re very likely not to find anybody there. I mean, my friends tell me, don’t even bother. You know, so that doesn’t mean the work week is shorter, but it means, you know, that there’s substantial changes in the real estate sector and, you know, excess capacity of one kind, people—economic activity is moving around.

I mean, on the whole, I would say the investment situation in the United States is reasonably healthy. You know, for the first time we have sort of major investments in infrastructure, you know, that have been funded by the government. And the CHIPS and Science Act has some more major investments, some of it designed to bring activity at home. And then we have the Inflation Reduction Act, which is designed to put, you know, funds into the energy transition in pursuit of sustainability. So when I look at the whole—I mean, there’s imbalances all over the place because of these structural transformations. And I’m sure we could find places where there’s significant deficits. But on the whole, I think the investment program, you know, or the investment situation looks moderately healthy. You are going to see just huge investments in the digital technology side as people pursue this set of opportunities.


BLOCK: Yeah, I don’t think I have anything to add. I mean, it is—it does feel like we are seeing more manufacturing jobs. I think we’re all—having come out of the Biden administration, I’m really excited to see sort of the full implementation of the Inflation Reduction Act and the CHIPS and Science Act. We just—I think, last week the president visited a site of the first—like, one of the first major investments. So I think that might balance out, you know, the kinds of trends that the questioner was raising.

FASKIANOS: Emily Walker, who’s legislative director for Pennsylvania Senator Katie Muth, asked: Can you talk a little bit about the wave of organizing that’s been taking place in southern United States recently?

BLOCK: Yeah, happy to. It’s a very interesting dynamic. So a couple of different trends. There has been a concerted effort, particularly driven by SEIU, Service Employees International Union, to do some kind of innovative organizing in the South. You know, the South is a very challenging place for the labor movement. Has been for a long time. And so there’s been a push to not do traditional union organizing but just try to get as many workers engaged in collective activity without necessarily using the traditional model of an NLRB election for majority exclusive representation within their workforce. You know, the South now is pretty much universally right to work, which just makes it a very challenging environment for traditional union organizing. So I think we’re going to continue to see these kinds of innovative campaigns. They’re really more like campaigns than organizing drives.

The counter to that, though, is, like in the Starbucks organizing, there’ve been about more than 300 Starbucks stores that have unionized, and a number of those are in southern states. The South has not been able to sort of put up that wall to union organizing, at least among the Starbucks organizing, that they have in terms of a lot of other sectors. But the other dynamic, which it’s too early to know whether it’s going to be successful or not, but is what I raised at the outset about the UAW’s intent now to organize the transplant car companies. Almost—not all of which, but which predominantly have located their manufacturing in the South.

And we also have—Ford is building the Blue Oval Plant, which is going to be, I think, one of the largest auto manufacturing plants in the country, if not the world. And they have now made a commitment to not try to stop the union from coming into Blue Oval. So that’s in Mississippi. That is going to be a union plant. That’s a big deal. But then the big question is going to be whether the UAW can organize other car companies that are not union in this. And I’ll note, they’re not union in this country. Most of these companies have unionized workers everywhere else in the world. And they seem to figure out a way to make money in plants in other countries with unionized workforces. They come here, and they fight the UAW sort of tooth and nail to keep the union out of their plants here, again, which are mostly located in the South.

So, you know, we’ll see. I think after this most recent UAW strike, underestimating the new president Shawn Fain is not a good idea. He did things in this strike that nobody thought he would be able to pull off. So I think, again, one of the big stories in 2024 is going to be whether we’re going to see inroads for labor in the South, particularly through these auto companies.

FASKIANOS: Thank you.

And I’m going to sneak in one last question from Charisse Childers, director for Arkansas Division of Workforce Services: Michael stated it is not possible to think that robots can operate on their own. Do we have employment data on jobs that were added solely in conjunction with added technology? In the same vein, jobs lost solely in conjunction with technology, meaning robots?

SPENCE: So, I mean, this a little bit nerdy, but, I mean, robot—human beings, you know, especially people who actually make things and, you know, do things in a physical environment, have, you know, an extraordinary capacity that robots don’t have. Which is an ability to absorb a rapidly evolving, you know, external environment, you know, visual and other signals, essentially with no latency. Robots aren’t even remotely close to that. And if you want evidence of it, look at the, you know, challenges facing the autonomous vehicles. You know, they do fine in highly structured environments, you know, where, you know, you’ve painted all the lines on the road, or you’re in a parking lot, or something like that. And then you put them in sort of an unusual situation, and they drive into a pile of cement or, you know, the emergency responders don’t know how to deal with them, and so on.

You know, in other words, in unstructured environments, you know, the robots basically need help navigating around, even if they have the mobility and, you know, manual dexterity, and other things that are other dimensions of robotics. There’s people working on this problem, but I think, you know, this is an example—you know, one of the many—in which I think robotics and people are going to work together. You know, and you’re not going to see full automation. Maybe in structured environments. I mean, you see some of it—you look at—it’s not just manufacturing. You look at a, you know, major distribution center, an Amazon distribution center, there’s—you know, there’s a lot of robots. And this isn’t very snazzy technology. They just don’t bump into each other and they go collect things and bring them to the people who pick and pack them, scan them, and so on.

So, you know, there’ll be progress in this. But my—having spent some time talking with AI people, I think that, you know, full automation, except in highly structured environments, is a fairly long way away. And we’re going to see mostly human—you know, human-machine kind of collaboration and those environments. And there are a lot of them. I mean, you know, if you go outside distribution centers and manufacturing things, and highly structured, you know, roadways and whatnot, pretty much everything else is unstructured, right? Hospitals, et cetera, so.

FASKIANOS: Thank you.

Well, unfortunately, we are out of time. But this was a terrific discussion. So thank you, Sharon Block and Michael Spence. We appreciate it. And to all of you, for your questions. We will be sending a link to the webinar recording and transcript, as well as some of the other resources that were mentioned. You can follow Dr. Spence’s work on CFR.org and Professor Block on X, formerly known as Twitter, at @SharBlock. 
And, as always, we encourage you to visit CFR.org, ForeignAffairs.com, and ThinkGlobalHealth.org for the latest developments and analysis on international trends and how they are affecting the United States.
Of course, please do share your suggestions with us for future webinars and any ideas on how we can help you in the work that you are doing in your communities. You can email [email protected]. We wish you all a happy holiday season. And we look forward to reconvening this series in fiscal year—or, actually 2024, which is right around the corner. So, again, thank you both. We really appreciate it.

SPENCE: Thank you. Thank you.


Top Stories on CFR

Immigration and Migration

Election 2024

Vice President Kamala Harris is seeking the 2024 Democratic presidential nomination in the wake of Joe Biden's exit from the race.


The closely watched elections on July 28 will determine whether incumbent President Nicolás Maduro wins a third term or allows a democratic transition.