Online Buddies for Jobseekers

- Job loss is common and often leads to a spell of unemployment. Usually that spell is fairly brief, measured in days, weeks, or a few months, but not always. Some job losers suffer a long period of unemployment, which results in a big blow to self-esteem, a lasting drop in living standards and serious hardship for the job losers family. Can a strong economy ensure that every job loser finds new work and quickly probably not? Is re-skilling plus a strong economy enough? There's certainly a role for re-skilling, but long-term unemployment is not just about skills and economic opportunity, it's also about the job seekers outlook, awareness of alternatives, and openness to new types of work. Welcome to Economics Applied a Hoover Institution podcast, and thanks for tuning in. I'm Steven Davis, a senior fellow and director of research at the Hoover Institution. With me today are two economists that I count as friends, Michelle Bilal and Philip Kerscher. Both are professors at Cornell University. Both are creative and highly skilled researchers with many insightful studies to their care. To their credit, they are married to each other and parents, so they're partners at home and sometimes in their research collaborations, as we will hear today. Welcome to both of you. It's great to have you on the show. Thanks, Chris. Thanks

- For having us. For having us.

- You know, one of the reasons I wanted to have you on the show, you have an interesting new study that we'll get to, but I, I've, I've spoken to you going back some years about how you think about long-term unemployment and joblessness, and I think you have a perspective that is somewhat different from, or maybe more expansive, let's say then what is often taught in traditional economics courses. So I, I wonder if maybe you can just start out by saying a few words about how you think about long-term unemployment and what, why some job losers get stuck in that state. Who, who wants to take that on?

- I'm happy to take it on. So I think there's several things coming together when you talk about long-term unemployed, right? You know, we did some studies a few years ago in the UK dedicated towards long-term unemployed. And what you see among the really long-term unemployed, by which I mean like a year and more out of a job, right? Many of them have disabilities that prevent them from having jobs. A number of them haven't really done real work before. They did some small things, but, but not much. And so, so these are some of the things that, that prevent them from having jobs. But the other thing that we see when we look also at people that are longer unemployed, but maybe more of the six months duration and so on, is, is people that are looking for work in areas where there's currently just not a lot of jobs out there, right? And, and for those, and that could be because of technological change and other changes and you know, reorienting themselves trying to figure out what could I do otherwise? What, what skills do I have? Or if I don't have them, which skills could I pick up on the new job? Right? Or where do I even have to invest and get more skills that, that seems to be lacking? And through the studies that we have done, we see that people seem to be somewhat narrow. Not everybody, but, but there seem to be people that are somewhat narrow and you wonder whether their job search strategy is really making the best out of what they got and whether that's might be why they are.

- So two, two things here. One, I guess maybe I should have prefaced all this by saying, even though most people don't get stuck in long-term unemployment situations, it is never the less a socially very corrosive economically and socially for their families, sometimes for their communities, certainly for themselves a corrosive phenomenon. So it's something we wanna, we want to minimize as much as possible. So I think that's, that's part of the backdrop too. You know, this conversation. The

- Thing I should maybe add to the previous that, that we said, is there also people, and that's something that economists have worried all about for a long time that might get discouraged.

- They get discouraged, right? That

- Yeah, they, they, they don't find something and they might not have the, the support network or the mental strengths and so on to keep up the type of job search that it would require to get them a job.

- Okay. So, so I hear you saying that there's, there's a psychological aspect to long-term unemployment. There's two aspects. One is the discouragement you just mentioned, but the other is this kind of, sometimes people need to reorient themselves and how they think about their own skills and their own potential role in the labor market that's distinct from discouragement. So they might keep applying for jobs and looking for jobs, but they're looking in the wrong place in some sense because what they used to do is no longer what's out there as an opportunity. Do I hear you correctly?

- That's, that's the biggest part that we try to tackle.

- Okay. And that, and that I think is a little different from the usual approach of economists, which is simply a, they kind of jump over this psychological aspect. They just say, okay, you used to do this with these skills, now you need to do this with these skills, give you a retraining program. But people have to be persuaded to change their outlook on what they're going to do in the labor market. And for many people, not for everybody, it's tied up with their, their sense of identity.

- Yeah. And actually it, we had a conversation, now it's more than 10 years ago in the UK with job seekers. And it, one of these conversations probably triggered the whole agenda because there was a lot of programs, you know, around retraining, but also around, you know, giving incentives to job seekers to search harder. It's a little bit, you know, the econ 1 0 1 kind of approach, you know, maybe you're not searching hard enough and so on. And this person told us, look, at some point you can give me a million pounds. I still don't know where to look. And, and we thought it was interesting, we said, well, what can we do to help people rethinking where they should apply?

- Okay. Okay. So great. That's, that's useful. And so let, one more piece of background to your study that I think is important to get on on the table is the role of social connections as a resource or a potential resource for job seekers. And this is an old idea in economics and in sociology, but I can you just expand on that because I think it's also a piece of what motivates the, the study we're going to talk about in detail.

- Yeah, so you're right that this has been, you know, for decades, sociologists, and perhaps a little bit later on, economists have actually noticed that social connections matter a lot in, in job search. That most people, a lot of people find their job through connections, but also we think social connections can be helpful not only to potentially identify a particular job, but also to help you navigate the process. Maybe understanding, you know, how how should you apply, how many applications should you send and maybe even help you psychologically to navigate the setbacks and so on. So this is something we know for a long time, but there hadn't been work where there was really a a, an attempt to create these connections to really, you know, leverage this knowledge that we have and try to design an intervention around it.

- Okay. So I, I heard you say three things about social connections. First is just information about what skills you need, where the jobs are, and that that's an aspect of social connections that economists have been all over. Then there's the process point. Well just how do you actually do it? You know, how do you navigate the process? You know, there's some work in economics on that, but maybe not enough that, you know, that that's actually a challenge for people, especially if you haven't even looked for a job for 10, 20 years. So it, it's a challenge and that kind of complexity in navigating a new decision making search process. There's some work in economics, but the last one you mentioned is, again, back to the psychological aspects where I think sociologists may be, maybe ethnographers get into that, but economists have, have not done a lot on that issue. Your, your work being, being something of an, and I don't mean just today, but going back, I, I remember having a conversation with you many years ago in which you were describing to me in another interview, it might have been in the uk and it was some guy who said, well, I'm a pipe fitter and I've been looking for pipe fitting jobs, but, you know, there weren't any anymore pipe fitting jobs. But that was how he saw himself. And the challenge in that case was to reorient him that, that that person to think about himself in a different way as he related to the labor market, if I got that right.

- Yes, you got that right. And I think it was actually, it was surprising how specific people can be, even within pipe setting, this particular person had a particular type of pipes that he was fitting, and already these other type of pipes were kind of outside of his right. So, so, so people have sometimes a very narrow outlook on both what they do, but maybe also in, maybe that's related to the last point that Michelle mentioned on, on who they are, right? So when you tell people that there are then other jobs out there, they also have to not only think about do they have the skills or could they learn it and so on, but also could they envision themselves in this skill? Could they envision, right. How they would be accepted in that type of job and so on. Right? And that might not be something that people are completely familiar with, especially when you think about transitions from, let's say more manufacturing into maybe more services and all of these kind of things that, that really involve quite different.

- Okay. You, you, you kind of we're kind of talking about the role of envisioning yourself in some new position, the labor market looking back. But this word resonates with, with the present moment a great deal. There are many people who are having trouble, I think, or anxiety about envisioning themselves in what might be a very different labor market in the coming years where AI technologies are transforming the nature might this, this is, this is not very far along at present, but have the potential to transform the nature of certain jobs. Maybe displace some, maybe alter the task mix in others, maybe require a different sort of skills. You hear technologists and others claim this is gonna be a tremendous transformation. I think some of that's probably overstated, at least as to the speed. But nonetheless, it's created a lot of anxiety and it's partly because people do find it challenging to envision how they will fit in to that new world. So this conversation isn't gonna be about explicitly about ai, but I'm kind of underscoring the point. 'cause implicitly it is a little bit about AI and what it might do to our, our, our economy and our labor markets.

- I mean, it is a, not, not completely unrelated in the sense that the, the sample we target is actually a sample of people who have challenges finding jobs. So are looking in labor markets where, you know, the number of vacancies is very small relative to the number of job seekers. And many of these jobs are the jobs we talk a lot about in the context of ai. Okay. Things like administrative assistance, right? You know, receptionist jobs that are in the sort of middle range of the, the wage distribution. And that have been talked a lot in, in sort of economics.

- Okay. Okay. Great. So one, one more background issue is you, you had the idea that employment agencies of the sort that I, that I think are especially public employment agencies that are perhaps more, more prevalent, they are more prevalent in Europe than in the United States. But think, so think about public employment agencies or employment agencies more generally might play a useful role in facilitating the types of social connections that, for the reasons we've been discussing, help people find new jobs, new even new career paths, new types of work. So explain that to us. What's the potential role of employment agencies that might help deepen social connections in a way that helps people find new paths in the labor market for themselves?

- So one of, one of the great resources that I think, or that we think these unemployment agencies have, but that is currently a little bit dormant or underutilized, is the fact that there are a lot of people going through them, right? They usually administer everybody who goes through a spell of unemployment and a lot of these people find a new job. And at that point, usually the connection breaks with the unemployment agency and everybody else,

- Right?

- But some of these people might be willing to interact with people that have yet not yet managed to find a job. So there's this this resource of, of a continuous flow of people that managed successfully to get out of it and might be willing to pay back to society by talking to and potentially engaging with people that haven't done that step. And that's what we wanted to utilize because they this potentially hundreds of thousands of people and even if only 10% of them volunteer, right? You have quite a, quite a number of people that that would be willing to interact. And that's what we try to utilize to create more connections to those that, that have not yet found a neutral.

- Okay? And so even though the, the, the, the exact nature of these employment agencies or unemployment agencies and how they function differs across countries. Okay? But I wanna point out for our audience, which presumably includes many Americans, even in the United States, there is, in every state in the United States, there's an unemployment insurance agency of some sort, often called employment security administration that does exactly what you describe for of people. There are millions of people who collect unemployment insurance benefits in every, any given year. They are tracked as a matter of course, while they're employed, when they become unemployed and apply for unemployment insurance, the benefits in the United States, and then after they get reemployed. So this administrative apparatus already exists in pretty much every advanced economy in the world. You're not talking about creating something new, new, some new bureaucracy, some new apparatus you're talking about. We already have this apparatus, why don't we make more effective use of it? And that's what I understand to be the kind of the, the, the core insight to use it to make these social connections, to help potentially help long-term unemployed or, or any unemployed people find new work, if I got that right.

- Yes. And I think o one thing to note is that of course we all have social connections and you think, well, why do we need, you know, these additional connections? But probably everyone, when they are thinking about, well, who do I know, how many, how many people do you really have in, in your network who have recently changed career and who could actually teach you something, it is probably quite limited. So this is what we think is actually useful and valuable to connect you to people who might actually have just made a transition and could

- Help you. Right. Okay. So let's get into your, the details of your study, by the way, which is titled Online Buddies for Job Seekers A Field Experiment. Okay, that's a, that's the title of the study. I think we've, we've now set up the motivation pretty clearly. I wanna get into the details, but first perhaps tell us what you mean by an online job, buddy. You've sort of alluded to it already, but exactly what do you mean in this context? How does the employment security agency help in this regard? How do you get people to play this online buddy role?

- All right, so the reason we called it online buddy, because the, the goal of from us was to make this as simple and as little effort for everybody involved as possible. In particular also including the employment agency because they have a lot of stuff on their plate already. They, they're not into right? They, they, they, so, so what we did is we created a platform where the, the buddies could register buddies are people that already found the job in a new occupation. We basically, through the unemployment agency, we had already done a previous study. We asked people in that study that had indicated that they're willing to be contacted again, whether they would like to be a buddy. We also advertised outside because we didn't have yet the huge amount of buddy. So people that are willing to make bo to to be buddies could register on this platform, give certain information, kind of create a little profile that the other side could see. And this platform also allowed the, the unemployed people, which we newly invited to take part in this, to register again with some little description of themselves. And so these online bodies are on that platform, we try to create a match and we could talk more about how we created that match in the end, right? And then on that platform, they could do the first steps of this interaction, right? First to find each other, to accept these matches and then to send messages through this platform or to basically schedule a meeting. And that meeting could be taken outside of the platform on phone or Zoom, whatever they wanted to do.

- Okay. - Right. And, and the online aspect is mainly about this bridging in a very low cost way. The, the two sides together.

- Okay. So the initial matching, and as I understand it, these buddies are volunteer volunteers. They're not compensated, they're not incentivized in any monetary fashion as I understand it. And you had a explain your thinking behind that decision to go that route, other than the obvious one, the cost savings route. But I think you had another motivation.

- Yeah, I mean actually the cost saving route is obvious, but it's true that for employment agencies, it's, if you add something that's gonna be costly, it's gonna be difficult. But also it's true that you are willing, you are looking for people who are willing to commit a bit of their time and who are sort of interestingly motivated in, in the idea of helping others.

- Yeah, we - Actually have many examples online in many different domains of, you know, community platforms and so on. So it's something that also people are used to and we thought, well, maybe we can try this in this context and yeah.

- Okay. So you're, you're, you're relying partly on just the good spiritedness of people who have just gone through might have what might have been a trying experience. They found their way to a, a good outcome. And some of those people will be of a mindset, well, I'd like to help somebody else make, go through the transition I just went through. That's, that's the idea as I understand it.

- Yeah. And I think it might resonate to your audience when you look around the US volunteering has always struck me as being one of the Yeah. The things that people are really heavily involved in. Right? That's right. So, so, so now you find something where people can actually use what they just experienced as potentially a challenging time to do something good with it. And hopefully that motivates some people to

- Yeah, I, I like that point a lot because it gets to the thing, look, we dealing with social and economic problems isn't just about the government doing everything. We, there's a lot, there's a lot that can be done from the ground up. And this match, this kind of initial matching that you described, there's a role for the public sector there, but everything else kind of flows from the individuals helping each other rather than the so-called benevolent hand of the government kind of in a top down fashion.

- Yeah, I mean that, that makes me think about, you know, a question that we have received is whether we are trying to replace, you know, caseworkers with, you know, bodies basically. And our view has always been, look, the caseworkers have so much, you know, so these are the people you would meet when you are collecting unemployment benefits and you try to help you in your search, but they have, first of all, a lot to do. So it's hard to actually ask them to also be, you know, psychological support and, you know, helping you thinking about your career very, they might have a few minutes, but that's not, that's not a lot. So we really see this as complimentary and I really like that you say that, you know, this is actually something where we can involve more people than just, you know, the public implementation.

- Yeah, my youngest brother was a, had a long career as a social caseworker and his general description to me of his job was, you know, border bordering on being overwhelmed all the time and having too many, too many people to help and not enough time to do it. So your comment about the, you know, helping, helping the caseworkers or filling in for what they don't have enough time, energy and to do is, it resonates with me. Alright, so let's get into the details of you conducted a field experiment. Okay? In other words, we've talked a lot about the motivation, the basic idea, but then you set out to do a scientific study in which I could, that would help you develop evidence, har hard evidence that would allow me and you, I should say, allow you to judge and me as a reader to judge the impact of the kind of online buddy system that, that we've been talking about. So explain lay, lay out to us how you set up this field experiment and why you designed it the way you did.

- Yeah, so we collaborated with the public employment agency in the Netherlands and we talked a lot about how we would actually design this. So this is really has been a joint effort that I want to emphasize. It's true that we wanted to have a, a scientific approach to this in a sense that we wanted to really have two groups of people, one group who would actually access the tool and the other who would continue, you know, business as usual. But in this setting you always worry about her. How do we actually really get a sample of people who are really comparable and one has access to the tool and one doesn't? So what we did is we recruited everyone and we told them not everyone will be able to access the tool at this, not immediately. And then we randomized, so we determined that random who actually had access to the tool, and that allows us to then later on for a a period over, you know, 18 months to be able to compare what happens to these people. And so the nice thing about the the, the, the Netherlands is that they were also, they're also really nice rich data that can tell us about what happens to, so whether they get a job, their earnings and so on. And so again, these are data that could be, you know, available in the US as well. But here it was really relatively easy. I mean, I put that between quotation mark, but relatively easy to match, you know, people who participate in our experiments to these kind of data. And so we were also able to conduct surveys with them, so collect additional data to see how did they use a tool and how they actually felt it, it helped, you know, their, their search. But so these are basically the, the, the, the approach that we take, we try to be a little bit like the heart scientist, but then in social sciences,

- Okay, so the unemployed job seekers who want to participate are randomized into two groups. Those who get immediate access to this matching tool that potentially puts them in contact with a online buddy who has previously gone through a similar job loss experience and found their way to a new job. And then there's another set of these people who, who are another set of the job seekers, the unemployed job seekers who are looking for a job. They, they're participating in your study, but they don't get access or don't get early access to the matching tool. Is that right? Exactly. You got the on the volunteer online buddies, right? And how did you contact those? You get the lists from how did you con how did you reach out to those people? And

- So we had already done an experiment with the unemployment agency in the Netherlands.

- I see.

- Before that was on people that were searching themselves in difficult occupations and some of them had found jobs in new occupations and during that experiment, right, you can't just recontact people if they don't want to, but some of them indicate that they're open to be recontacted. And so we recontacted them. We also simply advertised on social media that we are looking for people who had just made a transition of who were willing to volunteer by interacting with other people about it. And for those people, because they were coming from, from an, an unknown source, we had evenings where we on Zoom kind of introduced our concept, what it is that people are not paid when they're buddies. So we really select people that really want to do this and, and, and selected some or had people sign up through that channel.

- Okay, so, and, and so how many of these job seekers were in the study? Both the, the ones that got access to the tool and the ones didn't? How many, I'm just trying to get some sense of the size of it. So

- We had about 260 in each group. So the idea was, right, so this is a medium scale study that is, that is in some sense a proof of concept, right? Right. Before you roll something like this out on a large scale, you kind of wanna know that this stuff somehow works. Right? Right. And, and so doing an initial study of this scale seemed to us have enough potential to detect effects without yet dedicating all the resources into it.

- I understood. And then the buddies, how many, how many potential online buddies were there, who came forth? And then how many actually participated?

- Okay, so, so we had about 50 bodies at the end who got connected to someone? How many? It was probably two or three times more actually at some point indicated, you know, interest but then may not have showed up at the, the zoom meeting or you know, we basically dropped along the way. So it's, so one thing that I think is, is really useful to point out that is, is that it is, obviously in this type of studies you always, you know, lose people along the way, but I think there's something really quite interesting about trying to test this on, on a small scale with the people who are, you know, potentially probably most motivated maybe, but at the end, if you, you know, if what you find is going to be, you know, telling you something about whether there is any promise in this tool at all.

- Yeah, well that's, that's kinda the, you you already anticipated where I was going, the reason I'm pressing you about these numbers is there, there is an issue which will be what could be addressed in a future study, which is how scalable is what you find. So you're operating at a, a scale that is large enough to draw inferences, scientific inferences, but, and, and provide a proof of concept. I think that's, that is the natural place to start. But, but we, we should understand there's still gonna be open, an open question, how much could this be scaled up? And I think a, a particular aspect of the scaling is how many of these online buddies would potentially come forth. Right,

- Exactly. And, and this study here, I mean it's, it's kind of on both sides, right? Because this study tells you the effect for job seekers that would actually welcome this, right? Right. These are job seekers that already anticipated that they would actually welcome such a meeting, right? The exact share of job seekers that would welcome this is something one has to figure out. And probably as in this study, the bottleneck is probably the number of buddies, but given that there's so many thousands of people going through there, I think could potentially be a good start.

- I mean, another thing is that I think it's true that one may think, well we are looking maybe at, at quite a selected sample of people who are willing to participate and so on. On the other hand, we are also starting cold. People don't know anything about this platform. Yeah. There is no reputation or anything. So I'm always hoping, you know, it's a bit what we know from technological innovation as well is that it's hard to start, but once people find out about the tool and might actually realize this is actually nice and useful, then maybe it does actually become easier to recruit people over there.

- Well that that's a good point. And clearly there's the explosion of social media is, is an, is a latent signal that people are trying to figure out how to navigate their lives through these connections that they create on social media. So I do think there's, there's potential there to make this, you know, to scale this up quite a bit. But let, let's, let's, we're, we're speculating a little bit here. I certainly am. So let tell us what you found in this particular, in your study, the one you've already executed.

- Right, so this study in terms of the effects that we found, we, we, we track people for 18 month for which we have pretty good data and we see that both employment and earnings go up. But you know, it, you know that that's not what we call statistically significant initially, right? You see something, but you wouldn't say that it's jumping out of you really, after about a year, these things accumulate to a point where you really see di these kind of significant differences in employment and earnings for kind of this, this last half year where we track them. And that's kind of just to put this, if you, if you add the overall earnings that they have more than a person from the control group, that's about $2,500 per participant in the treatment group if

- You're per year. Per year

- Over these 18 months.

- Oh, over the

- 18

- Months. Okay.

- Over the 18. And so what is kind of interesting to us is why did it start to become significant kind of late, right? And so when, when you do a study like this, you usually write a pre-analysis plan where you lay out in advance which, which things you're gonna study. And one of the things that we had pre-specified is that we look separately at people that were already a little bit longer unemployed. Some people enter our study, they are pretty fresh, some people that are longer unemployed where longer unemployed is above four months in, in, in our specification. And what is interesting, the people that are longer unemployed for them, the effects kick in basically immediately, right? So basically already after two months, their employment is a lot higher and it stays higher throughout

- The, just to make sure the ones who got access to the tool

- Right.

- Among

- The long Exactly. So now we are looking at long-term unemployed, some of them got access to the tool, you compare that to other long-term unemployed that didn't get access to the tool. And for them you see that the, the employment effects jump up pretty immediately, two, three months into the study they have a lot more employment, like 9% of points more employment that goes throughout. And if you add up their additional earnings, it's about close to $6,000. Okay. That's over the course. That's, that's big. And that's about, that's about a third more in earnings and that's nearly completely explained by them working a third more of hours. So it's not really, they get more per hour or something, they simply find more hours to work.

- Okay. So that's, that's the proof of concept.

- Yes. - That, and I'll just restate it as I understand it, for a group of people who are already pretty far along in their unemployment spell, they've been unemployed at least four months before they get access to this tool and they are willing to participate. They've, they've, they've kind of stepped forth that, that's the selection point we discussed earlier. This buddy system really does help 'em increase their earnings quite a bit. 6,000 you said that's like a one third increase over the baseline earnings for this group. Did I hear you correctly?

- Yeah, they were all unemployed at the, that's what we, that we condition on. Right? They all had no, no earnings at the point where we gave the intervention and yeah, they, they earn about $6,000 more over the course of

- On average in people in the control group.

- Yes.

- Okay. So that, that's big. That's, that's quite a success. Now this is, again, this is a, a moderate, moderate scale study, but that, that's a big effect and it does suggest if this could be scaled up effectively, there's, there's a lot of potential gain to be had here and it's not costing, I mean there's some administrative cost to operate this, but this is not the, the government's not paying these online buddies. In fact, it's gonna collect tax revenues on the extra earnings that this, so this is like a positive, this is a positive NPV under finding, if you want to think about it that way. In, in a pretty big, at a pretty big level.

- That's what we think of.

- Okay. So what do you think is happening now? We, we, earlier at the outset of our discussion when we talked I think about three different ways, three different potential reasons for how social connections could improve the, the job ex the job search process information, navigating the process and dealing with this kind of psychological barriers to reimagining yourself in a different role, which do you have some insight in on the relative importance of those three?

- Yeah, so we actually collected survey information through surveys. So there is going to be based on the people you know who take part, but I think we had a number of questions to get at these different mechanisms including, you know, things like asking them how, you know, optimistic they are about, you know, searching about the outside of their, their careers sort of usual career. We are asking them about emotions as well. So all sorts of things together at this mechanism as such, we see that people seem to be indeed applying more, you know, to sort of different occupations. So when they can sort of, they're asked to list, you know, which occupations are you interested in? And we can see that the treatment group seems to be listing more things that are outside their normal occupations. And then we also see, I think what's interesting is in the survey that is more about subjective, you know, subjective beliefs or views about different things. We see that they seem to be lowering a little bit their expectations about the salary they might get when they restart a job without experience. And this we think is quite interesting because again, we know there might be a lot of anchoring going on where you sort of lost your job, you think about the wage that you, you used to have. And so maybe talking to someone else might help you realizing that maybe you will not earn immediately that same wage again, but you might need, you know, to first build, build some experience again in a new job to, to increase your earnings. And I think that's, that's sort of the main margins we see movement in.

- And you see that also just to follow up, you see that also on the dynamic component of this, right? So the control group seems to significantly decrease the number of occupations that they list over time.

- Hmm.

- Right. As a, as a dynamic process. Hmm. Also the breadth, we have a measure of how diverse are these occupations that also shrinks and that's not true for our control groups. So maybe just to, to kind of right, they, they seem to maintain this broad outlook or I see over a longer course of time.

- I see. So it's interesting that the role of the buddies here, so o obviously, you know, the, the kind of simple minded economist approach would be just tell these unemployed job seekers, they should look in other occupations and oh by the way, here's a list of places you could look and we could certainly tell them you should expect maybe you might have to take an earnings cut if you're gonna switch operations occupations at least at the outset. So that, that's the information piece. And employment security agencies I think have long done that kind of thing. Certainly telling people about where the, where the new job opportunities are, but that doesn't quite cut it for many of these people. It's actually the interaction with another human being who has gone through a similar experience recently. That seems to be important here is what I'm hearing.

- Yeah, I mean I think this is interesting, you, you know, in other domains as well that there is always something interesting about thinking about what are the peer, the peer modeling, you know, approaches doing. And it's true. You could say, well it's just, if it's just information, let's just give that information. But there's maybe something about the human relatability that that makes this information maybe more credible that maybe you retain better that information, which is interesting. But this, we are speculating in a sense that we did not really, we did not want to actually micromanage all the conversations and exchanges they had. So we have very, very little information about what actually went on in these conversations a little bit. And so we see this through the surveys, but it's, it's, we are speculating

- It's also hard to, it's also hard to say Yet whether there is not a way of transmitting some of that information even without the personal connection. Yeah, right. For sure. So a part of our platform we did have the ability to watch also videos of people that have gone through this process that basically lay a little bit out about whether they were a good fit and so on. We, we, we know what these people said, so it's not about the fine details about how you get a job, but more how do they ma how do they felt in the transition, how did they manage, how do they feel now that they are in it, whether they would do it again, things like that. Right? So there were some of these that, that people could do digitally either while they were waiting for a body or if they couldn't find a body or just if they were interested. And our study is not designed to distinguish kind of how big an impact you could achieve by transmitting this information in a different way. Right. I think it's also still something that's kind of interesting. What, what I think doesn't work is to just do that out of the cold. These people are already now in an environment where they expect to be matched with the body, where they're open to this, where they want this information. I don't think you can just mail it to them and we actually know that because

- Right.

- Yeah.

- Okay. Yeah, maybe your study is, your sample size isn't large enough to get at these kind of questions, but did you learn anything? Is there anything to learn about the characteristics of buddies that facilitates better outcomes or about the similarity or dissimilarity of the, the buddy and the person they're matched to? Is there anything you can say about how on either of those dimensions, how it led to more effective or less effective relate outcomes?

- Yeah, I mean this is a very good question and of course it is probably, you know, relevant the only heterogeneity that we had pre-registered before and that we look at, but it's actually not great because we are getting into very small sample sizes. Yeah. Is how close the body is to you in terms of the occupation you were actually looking for. Yeah. And so that, that is an obvious one. We have some suggestive evidence, it's not statistically significant so I want to make you know, too much out of it, but it is, goes in the direction you would expect that of course if the body is closer to you, it seems a bit, you know, more, more effective. But of course that would be great to actually have more information about, you know, matching on, you know, gender, all sorts of, you know, characteristics that you could think of that might matter

- With a lar with a larger study you could optimize, so to speak the buddy matching process and probably get even better outcomes

- And that you could even do if you had a study where the individuals are not randomized if, if you just imagine this actually were ever to get rolled out by the employment agency on large scale, you could at least still figure out by matching these bodies a little bit randomly which association of body and job seeker might really work.

- Right, right. So let, let me wrap up with a couple more questions. One, so you carried out this study in the Netherlands. Any plans to carry out a similar study in the United States?

- Well, I mean, so again, great question. So we are indeed very actively trying to, to do things in the us So we have started something, if I can advertise or Yes please. Job search lab. So we have created, you know, a little lab here where one p key piece of, of this lab is actually what we call an experimental job search platform. So we've tried to create our own, you know, job search platform on which we have control and we can actually then use it to do various, various things. And so this is where, you know, we are going and there might be indeed projects we can do that follow up on what we did in the Netherlands at the moment we have done a a a, another sort of large project in, in New York state that relates to more sort of giving people information about how they compare to other people who seem to be searching for similar jobs. But, so we haven't really implemented the body sort of setting here. So

- What about, have you, are you, have you sought or are you going to seek to enter into a cooperative relationship with any of the state employment security agencies in the United States? Many of them have serious funding problems. I'm sitting in California, they're massively underfunded, meaning they borrowed from the federal government because the, the, the payroll taxes that are collected that feed into the unemployment insurance system haven't been large enough in recent years to cover the unemployment benefit outlays. So there's, you'd think there'd be a, an appetite for we can, we can help get people back on employed sooner and that means less of a drain on the unemployment insurance fund. Have you made that pitch to any, any states in the United States? And if so, what was the response?

- No, so, so we are trying to do this pitch and we do have, you know, some, some response. I do think that the way we proceeded in the past in, in, in Europe was actually starting with small scale studies and then showing, you know, what happens and, and basically understanding, you know, the context and will that work in this context. And so the idea with the experimental job search platform is really to go in that direction to do, you know, for smaller projects and pitch them and see if we can roll them out. Yeah.

- And so far right, there has been some collaboration on data to assess these studies, right. But we haven't yet made the large pitch in getting employment agencies from the US to do that. I I do hope that that will materialize. In Europe we see a lot of reaction to these type of studies. You had organized a large conference with a la a number of these national employment agencies from the Netherlands and Germany and so on. And, and you see the, the, the interest in going into this space and also to, to not only work with private sector companies where the, the knowledge is usually quite proprietary right about, but also working with with people that are in the, in the research area to see whether there's something transferrable that can Right. That, that can really aid. Right. And, and so we hope that the same interest will maybe pop up here.

- Actually its a question you had asked me a few years ago, like why, why is the market not doing this? You know, and, and sort of in general for many, you know, interventions we can think of, it's always important to think, well why is the market not doing this? And of course there are platforms that could potentially do these things. No link people where we know they are

- Right.

- Platforms like that and, and, but yeah,

- Well the market, there may be some, you know, some of this does happen, it happens informally on social media and, but, but for the reasons you talked about earlier, because these job loss events are, for many people somewhat rare and they don't have this experience maybe within their existing social network that they could easily tap into. There is the potential for an outside party, the buddy, like a platform you created or a buddy system in your study to facilitate that kind of social connection. So there, there, there is a, a natural reason why the market might not do enough of this from a, from a societal perspective. But I cut you off there Michelle. Go go ahead. No,

- I I I was going to say that, you know, if you take platforms like, you know, LinkedIn or Indeed or ser.com and so they probably have a lot of this, you know, information they could leverage on as well and that could potentially work. I I just don't know if you will have the same willingness to help, you know, if you're contacted by LinkedIn to try to help, you know, maybe, maybe compensation then might, might become relevant, I don't know, but

- Yeah, but your, I think your, your instinct that motivating the buddies here through explicit compensation is probably not the right way to go. We want people who have a genuine intrinsic desire to help somebody who's in a situation that they just dealt with. And, and many people have that kind of goodwill. You know, if you know something that you think can help somebody else and you know, some people, not everybody, but some people will want to do that. So I think it's, it, it's great to tap into that. That's right. Anything else you wanna, anything else you wanna leave us with future research or on this line or just anything important we've left out?

- No, I don't think so. If people wanna know more about this job search lab, we have a website called Cornell Job Search Lab and there is not only our studies but also studies from other people on there and so on. And that's where, where we try to collect what, what many research groups try to achieve in this space.

- Okay. And we'll, we'll definitely put the link to that your Cornell Research lab, so research lab on the webpage for this program at the Hoover Institution. So if anyone's listening to this on some other platform than the Hoover Institution that go to the Hoover Institution page for this podcast, this episode, you'll find that link. We'll definitely put that up there.

- Yeah. And if there is anyone who is listening and think that they might be actually be able to help out implementing anything like that in the US we are also happy to

- Hear

- From you.

- Definitely. And if you get contacted on those one, I want to hear about it.

- Okay, sure.

- Wonderful.

- Alright, thank you. Hey, it's been great chatting with you. A real pleasure. And, and and I, I, I meant it sincerely. You're doing a lot of creative work in this area and keep it up. It's good stuff.

- Thanks

- Steve. Thank you so much Steve.

- Thanks for having us on the field.

- Alright, bye. Take, take care. Bye bye-Bye bye.