Transcript – Helping the over-50s find work

Bas van der Klaauw interviewed by Tim Phillips

Bas van der Klaauw 00:01

Once you have worked for a long time, you climb the job ladder, you climb the career ladder within the firm. When you lose your job, you fall off. And then the question is, where do you step on this job ladder again?

Tim Phillips  00:12

If you’re older and you lose your job, you might have a long wait to get another one. Policymakers often assume that job search training works only for the young, but my guest today thinks they got that wrong. Today on Voxtalks Economics, improving job search skills for the over 50s. Welcome to VoxTalks Economics from the Center for Economic Policy Research. I’m Tim Phillips. If you’re forced to look for a new job, it is a frustrating and stressful time, and the older you get, the more likely you are to spend a sustained time out of work. Often, young job seekers get training in how to navigate this process. If we give training to the over 50s too, would it work, and how would it work? Bas van der Klaauw of the Tinbergen Institute and CEPR helped run an experiment to test whether a Dutch job search training programme actually made a difference. He joins me now to share the results. Bas, welcome to Vox Talks Economics. 

Bas van der Klaauw 01:24

Hello, Tim. Thanks for having me here. 

Tim Phillips  01:30

Bas, if you are over 50 and you lose your job, how difficult is it to return to the workforce?

Bas van der Klaauw 01:37

So, it is much harder for the older workers than for young workers to find work. So, often it is said that their job finding rates are half that of the younger people, and it might be that — at least in the period that we were studying — over 50% of the older job search workers might become long-term unemployed. And that is not common for only the Netherlands, which we study, but also in other countries we see the same pattern. Older workers systematically have much lower job finding probabilities than the younger ones.

Tim Phillips  02:06

What’s going wrong here? Is it that they just don’t have the opportunities? They don’t know how to look for a job? Are they not motivated to find new jobs?

Bas van der Klaauw 02:14

I wouldn’t be able to pinpoint it to one particular reason. I think there’s a combination of common practice in institutions. So, typically the institutions are favoured towards the older workers. They have often worked for a long period, therefore are entitled longer to benefits. Benefits might be dependent on their last earnings. Older workers might have earned a bit because they climbed the job ladder a bit further. So, that means that their unemployment insurance is quite generous compared to the younger workers. Which also makes it harder to find a job again, which was similar to the job which they had before. At the same time, it’s not really like a common practice that older workers start a new career. Younger workers start a career, companies they like to have a younger worker who will grow within the firm, while an older worker only comes from the limited duration, which they still have until their retirement. So, they have a shorter period to develop themselves within the firm. So, HR people typically like the younger people more than the older people. So, there’s both factors which affect that it’s harder for older workers to find work again.

Tim Phillips  03:22

All over Europe, you see politicians describing their policies that are targeted at getting young people into the workforce. It’s a big political story in the UK as well. We hear less about the policies that they might create for this group, the over 50s. Why, in policy terms, is there less of a focus here?

Bas van der Klaauw 03:45

If a younger worker doesn’t find a job, you lose a full career. You lose many, many years of productivity. In general, we’re talking about older people in the labour force. It’s more about health issues; it’s more about keeping them fit until retirement. And it’s much less about getting them back to work. So, for older workers there’s more talk about, okay, when do you stop working, how do you stop working, what is the process towards retirement? For young people, the focus is always on getting them to a job, helping them start a career, keeping them productive, training them on the job. So, there’s always like this focus on work, while for the older ones the focus starts to shift a bit.

Intermission 04:29

Even when you keep your job, getting older at work presents challenges. In 2025, we spoke to Rita Ginja about the consequences of the menopause for women’s earnings in careers. Listen to the episode, The Menopause Penalty at Work, from March 2025 wherever you get your podcasts.

Tim Phillips  05:02

The experiment that you’re reporting took place in the Netherlands. It was a Dutch government program called STEP. So, what was STEP? What were its goals?

Bas van der Klaauw 05:13

STEP was a job search assistance programme, which was developed during a period that the labour market was in a recession. And the goal of STEP was basically to help older workers get familiar with job search skills that they might not have used previously. So, the idea was more or less like, okay, in this recession, older workers were also starting to lose their jobs. Often, they had worked for a long time. The way they applied for their jobs was from — let’s say, 10 years ago, 15 years ago — when maybe the digitalisation was a bit different. The way that people applied for work was different. So, the idea was to create a programme which would focus on more novel job search skills, and at the same time activate the social network of the unemployed workers. Participants would use their social network, and via their social network, would find work. It was very intense programme with 10 group meetings, two individual meetings. So, that’s much more intense than many of the other job search assistance programs. But the goal was really to get the older workers activated, even in a period which was not very beneficial for them. A period where more people were losing their job than finding work. A period where unemployment rates were increasing, and a period where that was basically the labour market was not in a really good shape.

Tim Phillips  06:31

And you evaluated this unusually using an experiment with treatment and control groups. So, who was taking part in this experiment, and how were they recruited into it?

Bas van der Klaauw 06:43

The programme was an expensive policy. The Ministry of Social Affairs was paying for that, but at the same time they wanted to have a proper evaluation of the cost and the benefits of this particular programme.

Tim Phillips  06:54

Okay.

Bas van der Klaauw 06:54

So, that’s why we set up an experiment, and the experiment took place around 2015, and it affected all the people at that moment who applied for unemployment insurance benefits, around that period in the Netherlands. They all took place in the experiment. So, it was not voluntary participation, everybody who was over 50 and applied for unemployment insurance within that particular time window, was either assigned to the treatment group or to the control group.

Tim Phillips  07:24

And if they’re in the treatment group, you mentioned it was a pretty intense programme. What are they actually doing day to day?

Bas van der Klaauw 07:29

Participation was voluntary to some extent, but the case worker would advise you and would kind of push you into the programme. If you were assigned to the control group, you were not informed about the programme, you had your meeting with the case worker. The case worker would say that you had to apply for work, but would not provide you with the opportunity to actually participate in this programme.

Tim Phillips  07:52

The people that were invited into it, did most of them take it up?

Bas van der Klaauw 07:54

Yeah, so we see that almost 60% of the people who were invited into the programme, actually participate in the programme. And we see in the control group that about 5% of the people ask themselves, can we participate or find some other way to participate in the programme. So, we see a little bit of participation in the control group and a lot of participation in treatment group, and that makes these groups different in the treatment intensity.

Tim Phillips  08:19

So, you have the people in the treatment group, they’re activating their social networks, which I guess, if you’ve been working for a long time, that might be an effective way to get out of unemployment. What was the effect? Did it help them to get out of unemployment?

Bas van der Klaauw 08:35

If we look a year after getting into unemployment — so a number of months after participating in the programme — we see that about 40% of the people in the control group, they have found a new job or they have left unemployment insurance. At the same time, what we see in the treatment group is that this effect is increased by about four percentage points. Which means that overall, a 10% increase in job finding due to the programme in the treatment group.

Tim Phillips  09:03

And do we know what the effect was on their earnings?

Bas van der Klaauw 09:06

So, they earn a bit more, but that’s basically because they find work faster, and if they find work faster, it increases their overall earnings. So, we see that there is an effect on earnings, which is roughly the same as the effect on unemployment insurance benefits. So, those two kind of things seem to cancel out against each other, so there’s not like an income effect for the participants.

Tim Phillips  09:27

As you say, the government was worried about the cost of the programme and whether it would pay for itself. What’s the news on that? Did it pay for itself?

Bas van der Klaauw 09:34

Yeah, so the programme is cost-effective for the unemployment insurance agency and the government. So, basically the reduction in how much unemployment insurance benefits were paid on average to the participant kind of outweighed the cost of programme participation.

Tim Phillips  09:50

This is good news, but is it helping some job seekers more than others? Who was it successful for?

Bas van der Klaauw 09:55

The programme was most successful for those people who were already like a bit employable. So, the ones who had a bit more education, those people benefited most from the programmes. The really low-educated people didn’t benefit that much.

Tim Phillips  10:09

And in the quality of jobs, I mean, one thing that’s always a fear for older people — like myself— losing a job is that you might never get a good quality job again. Are they getting jobs that were of similar status to the ones they lost?

Bas van der Klaauw 10:22

That’s difficult to assess. So, what you see, indeed, and that’s what you point out, once you have worked for a long time, you climb the job ladder, you climb the career ladder within the firm. When you lose your job, you fall off. And then the question is, where do you step on this job ladder again?

Bas van der Klaauw 10:36

Typically, a bit lower than where you were before. So, we do see that people do not manage to get, they take a job or unemployment causes an earnings loss to the individuals, but what we see is that the programme actually helps. So, the program helps people to find a job that gives a permanent contract. A job which has slightly more working hours. So, the quality of the job is improved by participating in the programme.

Tim Phillips  11:02

As you say, it’s a pretty intensive programme, and involves some face-to-face contact with people who are helping to deliver it. What do we know about the effectiveness of a good trader? Do we know what works in that respect?

Bas van der Klaauw 11:19

The whole programme has 12 meetings, of which 10 are group meetings, two are individual meetings, and these meetings are typically long. It’s not short meetings, like a half an hour, but it’s typically multiple hours or part of a day at which the meeting takes place. So, the trainer is actually important for what happens. So, if we compare different trainers with each other, and see, like, okay, what is the effectiveness of the programme if you’re assigned to the one trainer or the other trainer. We see significant differences between the programme effectiveness between the trainers. When we try to see, okay, what are the characteristics of the good trainers, we don’t find anything which we can really pinpoint. So, let’s say, if we think about gender or the age of the trainer, or even the experience of the trainer. We do not find that, for example, female trainers are better than male trainers, or older trainers are better than younger trainers, or more experienced are better than the unexperienced. So, we cannot really characterise what makes a successful trainer, but we do see that there are differences between trainers.

Tim Phillips  12:21

That’s really interesting to know that there are some trainers that are really able to make this work. We don’t quite know what it is yet. That’s always a challenge when scaling up, isn’t it?

Bas van der Klaauw 12:30

Yeah, it’s a bit unfortunate that we cannot really say, like, okay, this makes a good trainer, or this makes a good trainer, but yeah, so people with all kinds of characteristics are capable of being a good trainer.

Tim Phillips  12:47

So, on this evidence, there might be more scope for these active labour market programmes for older workers than many policy makers assume.

Bas van der Klaauw 12:58

Yes, I think we should be able to push older people back into the labour market or push or help them find work again. And at the same time, I must say that this was a costly programme, and it was evaluated during a period of recession, where there were many unemployed people, and they were at a high risk of becoming unemployed for a very long time. At least we now know that it works in a recession. In a period where the labour market is better, the question is, does it work as well? So, when it’s easier for everybody to find work, should we really help people with an expensive programme? Because those people who are capable of finding work, want to work, they should be able to find work outside the recession. So, it might be a bit recession specific that we here find that this programme is so successful.

Tim Phillips  13:43

And it’s also interesting to hear your description of it. It doesn’t sound much like a lot of the job search programmes that we’ve discussed on VoxTalks economics before, which are typically aimed at younger people trying to get their first job or coming out of university. What do we know about how to structure these programmes for older workers?

Bas van der Klaauw 14:05

So, what I’ve learned from this programme is that older workers use the same skills as the younger workers. So, to find work, basically the older workers and the younger workers use the same techniques, the same paths, the same methods to find work. Young workers come from school. They typically know these kind of ways to find work. The older workers, they have been working for a long time. They typically didn’t apply for work in, let’s say, 10-15 years, and now they have to start. So, they miss a bit of digitalisation, they miss a bit of information. It’s not that common that when older workers are unemployed, where young people, people typically start to look around, like, how can we help them? For older workers, you have, much less, so they need to put in a bit more effort into activating their social network, going in and asking, like, okay, do you have something, do you know about where your job is? So, the way that they search for jobs is not that different, but for younger workers, it goes more naturally than for the older. So, that is basically what this programme aimed for, trying to give the older workers the job search methods and the skills that the younger people were already using.

Tim Phillips  15:12

As you pointed out, this was during a recession in a Dutch economy, it was 10 years ago. Now a lot’s changed, but we are now in a situation where a lot of people are worried that they might lose their jobs and they might never get back into the workforce again. Do these different conditions now, and does the impact of AI, change how effective a programme like this might be in 2026?

Bas van der Klaauw 15:40

Yeah, so I think it might be. So, what this programme does is that it tries to help people who were unemployed due to frictional unemployment. So what happened in a recession is simply people were losing their jobs, not because they were unproductive, not because they were lacking skills, but simply because firms had to downsize. Firms were getting bankrupt, the changing conditions of work, like AI, that simply means that affects the productivity of people. That might be that people simply lack the skills to be sufficiently productive to continue working. And to fight this kind of structural unemployment, you need different programmes, you need AI training, or other kind of stuff where people lack the skills which they need to keep their job or to be effective in another job. So, I think if you aim for those kind of changes, you need different programmes, you need more training or more educational type of programmes rather than job search assistance programmes.

Tim Phillips  16:36

And if we don’t do this, if we don’t grasp the opportunity to get people in their 50s back into work, what do we lose as a society?

Bas van der Klaauw 16:46

So, we lose a lot. First of all, what you see over the past years, the past decades, is that it became much more normal to continue at working at an older age. You see, in many countries, that retirement ages are increasing. It is much more normal that people work for a longer period. And that’s also necessary, we have aging populations, we have more older people, and basically to keep the balance between the active and the inactive, we need that people work longer. So that is important for the public finance of the government. That also makes it easier for older workers to go back in the labour market. If it’s more normal that older workers work, it would be more normal for them to simply hire again an older worker. So, I think that is the importance for society is simply that we need people to work longer, so if they lose their job, we need to help them. We need the institutions and the programmes to get them back into another job where they can still enjoy working until they retire.

Tim Phillips  17:53

That’s huge challenges for policy makers. Thank you very much for talking about it, Bas.

Bas van der Klaauw 17:57

Thank you, Tim.

Tim Phillips  18:06

The discussion paper is called ‘A Randomized Experiment on Improving Job Search Skills of Older Unemployed Workers’, authors Nynke de Groot and Bas van der Klaauw. It is discussion paper 21464 at CEPR, if you want to find it that way. I looked up when my dad retired the other day, and I found out that he had already retired when he was my age. I hopefully have got a bit longer to work.

Outro  18:38

VoxTalks Economics is a Talk Normal production. The assistant producer is Megan Bieber, and our editor is Andrei Zagarion. Next time on VoxTalks Economics, from the Paris School of Economics CEPR Policy Forum 2026, Gita Gopinath and Philip Lane on Global Imbalances.