The staffing industry is facing a shift unlike anything we’ve seen before. Productivity is growing, but companies are relying on fewer workers—especially temporary staff. What may seem like a temporary market dip is actually the result of bigger forces at play: policy reform, immigration shifts, and rapid advances in technology. And temporary staffing, long viewed as a lead indicator for the economy and its strength, is now in a decades-long slump, even as overall employment remains strong. With these changes unfolding for the foreseeable future, the runway for forward-thinking agencies to build more resilient businesses has never been greater.
In this episode of Avionté: Digital Edge, host Chris Ryan, Chief Marketing & Strategy Officer at Avionté, unpacks these trends with Dr. Gad Levanon, Ph.D., Chief Economist at the Burning Glass Institute. Together, they explore why these shifts create as many opportunities as they do challenges—and why staffing companies that move beyond temp placements into broader workforce solutions, with a focus on AI-driven efficiency, will be the ones that thrive.
Key Takeaways:
This is a partial transcript of the full conversation. Listen to the podcast episode for the complete discussion.
Christopher Ryan: Gad, tell us a little bit about yourself and Burning Glass Institute.
Gad Levanon: The Burning Glass Institute is a think tank that studies labor markets, higher education, skills, and everything in between. I co-founded it with Matt Sigelman three and a half years ago. And, before that, I was working at The Conference Board for many years.
Christopher Ryan: So essentially, you’ve been studying labor economics and human capital for a long time. This is your primary focus.
Gad Levanon: Yes, it is. And it’s been a very interesting period to study labor markets.
Christopher Ryan: That’s definitely the case. So, Gad, let’s talk a little bit about the labor shortage paradox. And it’s funny, given some of the weaknesses that we’ve observed recently in the job reports.
You’ve been researching labor shortages in the US economy for years, so I’m kind of curious: is the problem that we don’t have enough people, or is it that we don’t have the right people?
Gad Levanon: Well, I think when we’re talking about labor shortages in recent years, it’s actually in the lower education occupations: blue-collar and manual services. That’s where we see most of the shortages.
And I would say it’s more that we don’t have the people who want to do those jobs. People who are retiring now in large numbers, many of them don’t have a college degree. And the people who are entering the labor market, a much higher share of them do have a college degree.
So, the workforce is becoming much more educated, and the number of people who don’t have a college degree is declining rapidly. And, if you get a college degree, you don’t want to work in blue-collar and manual service jobs. So, what we have is not enough workers in those occupations or people who are willing to work in those occupations. So, we have a labor shortage there.
On the other hand, in many of the professional and office jobs, we have kind of a glut of workers, which is now becoming even bigger because of AI and the way it can automate entry-level, professional jobs.
Christopher Ryan: I see. So, when we talk about jobs, I noticed that the plumber who came to my house a month ago, he was older than I am. He was in his late sixties. He could have retired immediately. And he had no one helping him. And he was very hard to schedule. And yet if I needed a lawyer, I could find one tomorrow.
Is that what we’re talking about? Is it skilled trades? Is it specific areas or professions, or is it all across the board?
Gad Levanon: First, I think we may have the same plumber. So, I guess it is a common thing. Skilled trade is definitely one of those things, because we had a very big decline in enrollment in community colleges, and a lot of the trades probably suffered from that.
But it’s also healthcare support, and even in construction, fixing, installation, and repair. Things like that. That’s where we see a lot of those shortages. Yeah.
Christopher Ryan: So, skilled craftsmen where experience is critical to the quality and speed of their work. The sort of thing where you need some education, but also almost an apprenticeship.
Gad Levanon: Yeah. And I would say among the more educated workforce, while there are less shortages in general, in areas like healthcare and social assistance, the demand for workers is growing very rapidly.
There have been some disadvantages because in many of those jobs, you cannot work remotely. That reduces its attractiveness. So, we may have some shortages there as well.
Christopher Ryan: If you wanted to have a career in the skilled trades? Are community colleges the primary place?
Is it an apprenticeship? I don’t see the Ivy Leagues developing a program for training electricians, plumbers, or HVAC engineers.
Gad Levanon: It’s still early days, but we may be seeing some recovery in enrollment in community colleges, especially in the trades.
But there are also strong efforts to increase apprenticeships in the country. So that could help. But I think the increase in educational attainment is still happening. More and more young people are getting a college degree, and they don’t want to be plumbers. So, we may still see some shortages in that.
I should add another thing. The new immigration policies, which are having a huge impact, especially on the number of undocumented workers in the US, are also hurting that segment. Perhaps in less very sophisticated trades, but in some of the more manual work in agriculture, construction, transportation, moving, and healthcare support. There are also increasing shortages in areas that have a lot of undocumented workers.
Christopher Ryan: Yeah. So, it’s a combination of both skilled trades and work that may seem less desirable to Americans, which we were looking to fill with immigrants.
So, I want to jump to questions around artificial intelligence and its impact on labor demand. You’ve written a little bit about this. I’ve seen this in some of your articles, and you have a unique visibility into how AI is changing job requirements.
I’m curious what types of roles are genuinely being eliminated because of artificial intelligence versus those that are being transformed? And where is AI creating new labor demand?
Gad Levanon: Well, we see a very striking impact on job ads or job postings for new entry-level or people without experience.
There is a very big decline there. Whereas in the more experienced segment, we don’t see any decline at all. So that’s a very striking impact. And when you think about it, what AI can’t replace is things like relationships, judgment, and experience.
But the more routine and technical things, like simple coding, are something that AI can do well. So that’s where we see a lot of impact, I would say, from all occupations. The coders are seeing the biggest impact. That’s like an area where AI is uniquely good at.
Christopher Ryan: So, everyone who’s been sent to coding school recently to learn how to code, potentially those are the people who are most at risk from artificial intelligence.
Gad Levanon: That’s exactly that. We’ve seen a very large increase in people who study computer science who are learning how to code. And that’s the profession that is now experiencing the largest decline.
Christopher Ryan: So, in some respects, the way I’ve been thinking about artificial intelligence is that it really favors experts. If you’re already an expert in your field, if you already have strong relationships, then you can leverage technology to actually expand your impact and be more effective in what you do. But if you’re starting and you need to learn the ropes, AI potentially is replacing you.
You may no longer have a career entry point. Is that the right way to think about AI?
Gad Levanon: In many cases, yes, it can replace things that you can learn in school, like how to code, but it’s not as good at many of the things that you don’t learn in school that are harder to define exactly.
Those are more based on judgment and experience. I think that’s an accurate definition.
Christopher Ryan: Got it. So, when we think about the weakness that was shown in the Bureau of Labor Statistics, in the BLS jobs report, of course, with the revisions that occurred over the last couple of months, to what extent is AI driving that versus other factors that might be in play in the economy?
Gad Levanon: I think we certainly have seen some slowdown in the economy, starting in April or March. Although I think April and May were the worst months. That’s where the sentiment shock was the strongest; almost everyone was talking about a recession. Stock prices went down. The sentiment was just terrible.
And I think since then, sentiment has improved. Companies are investing a lot. We are seeing some softness in consumer spending, but it’s still growing. So those definitely have an impact on demand for goods and services.
But I think that, in general, where AI comes in is that we are seeing relatively strong productivity growth. It started even before the pandemic. It accelerated after the pandemic. The big labor shortage that we saw in 2022 and 2023 really forced companies to try to learn how to do more with less. And now AI is coming in, you see a lot of layoffs in tech, and also some in finance and other industries, and reduced hiring.
So, I think all of those things together helped in slowing down job growth, because when productivity is stronger, you don’t need to hire as many people.
And on top of that, a very big impact came from, again, the immigration policy. There are probably hundreds of thousands of people who would’ve come across the border and didn’t, who would have stayed in work, and now they’re afraid to get out of their houses, so they don’t show up.
Many are voluntarily exiting the US. So, that also drove down job growth.
Christopher Ryan: Got it. So, let’s switch gears a little and talk about the contingent workforce, an area that’s of great interest to our listeners. The common wisdom is that the opportunities for contingent work, whether it comes through traditional staffing or through gig work or some other form, the common wisdom is that it’s going to continue to grow significantly.
Do you share that belief?
Gad Levanon: Well, as you know, much better than I do, that the temp health part of the contingent workforce is actually on a significant decline. And there may be other parts of the contingent workforce that are growing.
I would think that the shift to remote work could help with that. If you don’t have to be in the office, then the next thought is that maybe you don’t need to be an employee. But I don’t know that there is strong evidence that in the contingent workforce in general, there has been a strong increase in recent years.
Christopher Ryan: Got it. So, of course, the decline in temporary staffing occurred even as overall employment was going up.
Is the decline in temp staffing an early warning signal for the economy, or do you think there are structural changes? I know that this is a very difficult question, and it’s hard to find data. I’m just curious if you have any thoughts.
Gad Levanon: Yeah. You know, for many years, I thought that the temp help industry was perhaps the best leading indicator of employment. It really performed very well. But in the last two to three years, it’s been declining the entire time, without seeing a decline in employment. So, I think it lost a little bit of its ability as a leading indicator.
And I think part of it is because there are also both cyclical and structural factors that led to the decline in temp help.
I think, definitely, it helps when there is a high churn in the economy, so then companies are using temp help more, and there’s been a huge decline in churn, in quit rates, during that time. In general, we’ve been in a period of strong labor shortages since, I would say, seven to eight years.
And, when that happens, more workers are converting to full-time jobs, and the employers are seeking full-time workers. So that probably helped a little.
I think manufacturing slowed down in the last several years. Temp help is quite concentrated in manufacturing. There may be some regulatory pressure, especially in California and other places where there is a risk of misclassification, and that pushes companies to go to full-time workers. Also, there’s been a lot of criticism of companies that don’t treat their workers well, and that pressures them to provide more full-time jobs.
So, there may be some more Structural or long-term changes that impacted that.
Christopher Ryan: So, you know, I was thinking about this, Gad. You’ve been talking about or predicting labor shortages for a long time. And you know, a look at the population pyramid in the United States suggests that, as our population ages, people were predicting that there would be greater and greater shortages, especially among hourly workers.
And one possibility that I wonder about is whether companies have been investing in automation, not just artificial intelligence, but also other forms of automation, simply to run a more efficient warehouse or a more efficient food processing plant. Whatever it is, people are investing in automation because hiring new workers is expensive and turnover has been high.
So, there’s a part of me that wonders if automation itself isn’t actually part of this change, or this movement towards more permanent employment. When we think about policy: One Big, Beautiful Bill, tariffs, and immigration reform, you would think that, if illegal workers were no longer available, there would still be demand for the work in question. So, I’m just curious if you see any of those forces in the last year actually impacting the labor market, especially for contingent work.
Gad Levanon: Well, there are some who are obeying the law and don’t use undocumented or illegal immigrants. But there are some that do. I suspect that the ones that do are significantly impacted by the recent immigration policy. It’s much harder for them.
Probably they’re seeing less employers using them. Also, the workers themselves – illegal immigrants – are probably less in supply. But it may be then an opportunity for staffing companies that are not relying on illegal immigrants. Perhaps there will be a bigger demand for them.
Christopher Ryan: So, let’s talk a little bit about measuring the contingent workforce. Obviously, temporary labor and temp staffing is only one part of the contingent workforce. But getting verifiable data on contingent labor is extremely difficult, and a lot of labor statistics don’t fully capture the workforce.
And so, I’m curious, speaking with you as an expert labor economist, from your research, do you have any sense of the real size and growth trajectory of contingent work? And, if not, how should we be measuring it?
Gad Levanon: So, it’s not something I focused on very much in the last couple of years, but there is tax information for non-employer, self-employed, and that gives a pretty accurate description there. And last time I’ve seen, most of the increase was in transportation-related, so the Ubers, the Lyfts of the world.
But, outside of transportation, I haven’t seen a huge increase, but it’s been a while since I looked at that data.
One project that we are planning to do at the Burning Glass Institute is to see if we can identify Gig economy or self-employed contractors to see trends. In general, there hasn’t been a huge increase in self-employment. So, I think it’s somewhat hard to find strong evidence that the contingent workers overall are rapidly increasing.
Christopher Ryan: Got it. Yeah, that’s interesting.
So, looking ahead three to five years at specific industries, what types of skills and roles do you think will be in the highest demand, and, if you’re a staffing agency or one of their clients, where would you be focusing your recruitment and development efforts?
Gad Levanon: I think the areas that are growing the fastest are in healthcare and social assistance. In fact, in recent months, it’s the only growing sector in the economy, or almost the only.
But, I would say, we talk a lot about AI destroying jobs, but AI will also create jobs, and there will be a lot of new goods and services coming out that will create jobs. There is a huge investment in infrastructure – data centers, servers, power, energy – that will be needed to let the AI world function, and there’s a huge increase in that investment. For example, in manufacturing structures. There are a lot of jobs related to that.
Large language models, in some ways, there will be a science in the same way as there is neurology. A lot of management to understand better how to manage and how to get the maximum out of workers. There is an entire industry about that. I think there will be an entire industry around how to get the maximum out of large language models. And also things related to cybersecurity and ethics, and implementation in workflows.
So, there are a lot of jobs that will come out there. So, look, I think AI is a big shock, probably a very long shock, that it’s not going to be just a matter of five or 10 years. I think, in the foreseeable future, AI will get better and better, and companies will use it more and more.
So, a lot of jobs will shrink or disappear. But in history, there were never, perhaps excluding the Great Depression, very long periods of high unemployment rates. So, the market balances itself. When companies make money out of AI, they tend to spend it.
But the benefits from AI are not evenly distributed. A lot of it goes to people who are already quite wealthy, and they’ll become even wealthier. So, a lot of spending on luxury items, and we’ll see even more spending and jobs related to that.
If you just look at the number of millionaires in high-tech hubs and count them, there’s been dramatic growth. And those people – they will buy second homes. They will go on more luxurious vacations. They will get more domestic help. The people who work inside homes, I think that number is increasing and will continue to increase.
Christopher Ryan: Yep. That does make a certain amount of sense.
So, stepping back for a moment, if you were running a staffing agency, if you’re thinking about everything that we’ve talked about: artificial intelligence, policy changes, skill mismatches, and the evolution of work itself, what would your advice be to somebody who’s in staffing or some form of workforce management?
What do you see as the biggest opportunities moving forward in the contingent workforce economy? And how would you position yourself to capture it?
Gad Levanon: Well, I can think of two things. One, many of the jobs in staffing companies could be partly done by AI. Actually, there are more and more studies that show that AI is actually pretty good at interviewing. And, of course, all the kinds of administrative stuff that are around recruiting workers could also be done by AI. And, I think many staffing companies are already taking advantage of that, but there could be a lot of efficiency gains.
So, while maybe demand will decline, or is already declining, if this long-term trend continues, costs will also decline. So, that’s one thing.
The second thing is, if the staffing industry is shrinking indeed, and that’s like a long-term trend, staffing companies are not just good at recruiting temp health. They’re just as good at recruiting other workforce solutions in general. So, shifting perhaps a little away from the temp help part of it to just help companies do a better job in recruiting their regular staff.
That, I think, is a skill that many staffing companies have, which is something that many staffing companies already do. So, yeah, I think the market for hiring workforce management is not shrinking. It’s just, perhaps, the temp help part of it.
Christopher Ryan: Yeah. And it’s interesting because, at Avionté Connect, we had actually talked about this, and what we are seeing as this potential evolution from temp staffing towards a more holistic workforce solutions type of company, where there is consulting, training, recruiting, and acting as an employer of record for compliance purposes. There are a lot of different ways to participate in this economy that may or may not be simply around temp staffing.
So, I absolutely agree with you that there are some real opportunities here for companies. What I also hear you saying though is that there’s a huge opportunity to leverage technology, and that if it’s done properly, conceivably, a staffing agency can generate more profit and be more profitable – even if they have fewer placements.
Gad Levanon: I think it’s probably one of the industries that can benefit the most from AI.
Christopher Ryan: Yeah, there’s no question. That’s definitely an area where I think the technologist at Avionté would strongly agree with your point of view. So, I warned you that I would ask you a question about Gary Becker and Human Capital, and I saved that for the end.
And, for listeners who weren’t aware of this, Gary Becker was an economist who originally coined the term human capital. And part of his theory basically said that people earn less early in their careers, but more later on, in part through a variety of different factors, not just skills, but also internal knowledge of how a corporation works, how to navigate through an organization, through creating and building relationships.
And Gary really focused on full-time work settings.
So, Gad, as you think about labor economics in the 21st century, and you think about AI and automation, I’m curious, how well do these theories of human capital hold up when a significant proportion of workers have contingent work status? And is it still possible to build human capital and create a human capital accumulation when AI is breaking down the traditional career ladder?
Gad Levanon: I was thinking a lot about this in the context of my profession. So, I can give you an example of that. If you want to become an economist today, there are probably two main ways you can do that. You can, after college, go to a grad school, and then, when you are done there, you can immediately become an economist.
And the second is to, after college, become a junior economist or research assistant, and then move up and become an economist. I think AI is significantly hurting the second channel; there will be fewer research assistants, fewer junior economists.
So, I think one of the things that we’ll see in that case is that more people will go directly to grad school. And so maybe one impact is an increase in enrollment in graduate school.
Second, we will have to change. The research assistant or the junior economists that we do hire – we probably will hire fewer of them – but we will ask them to immediately do things that are higher skill. So, maybe we will ask less of them to do things that AI can do, like data cleaning and simple analysis, and we’ll just pick the ones who can immediately talk to clients or give presentations from the beginning. That probably will happen in a lot of other occupations as well.
Christopher Ryan: That almost sounds like a sink or swim scenario. You throw the junior economists in the water, and those who can swim are going to be the ones who succeed. And those who haven’t developed the appropriate communication or social skills or specific business problem-solving skills, because they haven’t been exposed, may not be able to continue in the same way.
Gad Levanon: Yeah. That may very well be the case.
Christopher Ryan: Very interesting. And it’s interesting that you would actually take that question back to the development of economists themselves. Historically, a PhD in economics was always considered to be a guarantee of income and employment for an entire career.
And it sounds like, even in the field of economics, artificial intelligence is a potential challenge for junior economists.
Gad Levanon: Yeah. And then, I would say also, partly because of the squeeze on universities, we see weaker demand for PhD economists, which is a troubling sign for the profession.
Christopher Ryan: So, what I hear you say ultimately, and what I’m taking away from all of this, is that the US workforce is under a significant transition and probably will be, not just for a few years, but for a few decades as we adjust to major changes, not just in policy, but in automation, artificial intelligence, and so on.
So, you are foreseeing almost a constant transition. It almost sounds like, if you’re in workforce solutions, you might want to specialize in helping people and employers make that transition?
Gad Levanon: Yeah, I think, there are big debates about whether AI is a big deal or not a big deal. I’m on the camp that thinks that it’s a very big deal, and not just a short-term big deal, but a decades-long big deal.
Christopher Ryan: So, this is going to be an economic period similar to the time in the United States after the Civil War, where there was continuous improvement in automation in manufacturing for 30 or 40 years. Is that what we’re looking at here?
Gad Levanon: I think, yes. I think that we’ll see continuous strong productivity growth in automation and transformation that will last decades.
Christopher Ryan: Got it. So, final question for you. If you are advising your children in their careers, knowing what you know, what would you tell them about what they should be doing to prepare to participate in this new economy?
Gad Levanon: Well, you know, my oldest daughter is a singer. She’s a college student, but she’s not the type of occupation with an especially strong workforce trajectory. But that is something that AI will be slow to replace. I can’t see a lot of people coming to listen to robots singing. So, maybe, in her case, it’s relatively improving her trajectory.
Christopher Ryan: Gad, this has been incredibly helpful. Any final words? Are there any trends that are keeping you up at night? Anything that you think our audience should be watching for?
Gad Levanon: Well, one thing we haven’t talked about is the impact on inequality. Gaps between rich and poor, and also regional inequality. Areas that will grow faster and areas that will grow slower. But there has been growth inequality for many decades, and I think AI will even accelerate that to some degree. But that’s an unfortunate topic for some other conversation.
Gad Levanon
Chief Economist, The Burning Glass Institute
Levanon’s research focuses on trends in US and global labor markets, the US economy, and their impact on employers. He is a regular contributor to Forbes and also often contributes opinion pieces to other national media outlets. In 2016, he won the Consensus Economics forecast accuracy award for the US economy for his work at The Conference Board. Levanon is also an adjunct professor in the economics department at NYU.
Christopher Ryan
Chief Strategy & Marketing Officer at Avionté
Christopher Ryan leads the Strategy and Marketing functions for Avionté. He brings more than three decades of consulting, thought leadership, and corporate experience in Human Capital Management.
Modern technology has revolutionized the way we live, work, and play. It’s also what’s fueling the gig economy which has dramatically changed employment practices. So, what does that mean for staffing and contingent work? In our Avionté Digital Edge podcast series, we will speak directly with industry experts to explore topics and trends related to the digital transformation of staffing and temporary employment in the US workforce.
