Killer evidence for intergenerational welfare dependency?

A joint post by Lindsey Macmillan and Ben Baumberg looks at an important – but easily misinterpreted – new paper on ‘Family Welfare Cultures’.

Iain Duncan SmithWith the topic of intergenerational worklessness high on the political agenda, a clever recent Norwegian paper on the role of family welfare cultures in intergenerational welfare dependency has been attracting some interest.  If you listen to the government rhetoric on the topic, the ‘curse of intergenerational worklessness’ is bred directly through families passing on welfare seeking behaviour from one generation to the next. This conjures up images of the (effectively non-existent) ‘never working family’ with the parent taking the child to the job centre on their 18th birthday to show them how to sign on. On the face of it, this recent paper fits into this rhetoric – but the actual message of the paper is more complex than it looks.

Evidence for intergenerational welfare dependency?

Prior to this new paper, there has been very limited evidence of parental welfare claims causing welfare claims in the next generation. The gold-standard review of the evidence over the last decade, mostly based on work from the US, Canada and Sweden, concludes that “while the intergenerational correlations in welfare receipt are clear, there is much less evidence that a causal relationship exists”.  In part, this is because identifying causation is difficult in an intergenerational context. But in part this is also because there could be many potential channels that could lead to a correlation in claimants across generations that are not driven by the parents’ receipt – one obvious example being exposure to similar local labour market conditions.

The new working paper by Dahl, Kostal and Mogstad (this seems to be the latest version – but bear in mind that it’s still labelled ‘preliminary version’) uses a neat mechanism for separating out the effect of parental welfare receipt. They use an anomaly in the Norwegian benefit system where borderline cases are randomly assigned to judges for review, who then decide whether claimants are entitled to welfare receipt. The authors compare borderline disability insurance (DI) cases assigned to a lenient judge vs. borderline DI cases assigned to a strict judge – the crucial point is that for the two groups, the only difference is that one got lucky with a lenient judge and those parents received DI. The other group wasn’t so lucky and did not receive DI.

When they compare the welfare receipt of the adult children on these parents, they find that those with parents who received DI were significantly more likely to claim this themselves than adult children with parents who were assigned a strict judge and therefore did not receive DI. This suggests that the DI welfare receipt of parents is causing their adult children to claim DI themselves.

Figure on intergenerational transmission - described in text

All is not what it seems

The headline results seem pretty strong evidence for the ‘dependency culture’ idea – yet when you go further into the data, you see a very different picture:

1)      The sample of data is restricted to parents with adult children at the time that the parents’ appeal decision is heard by a judge. These aren’t children growing up in a home where DI is received throughout childhood. In fact, they are at least 18 and likely older at the time the parent starts claiming DI (the key results are just as strong if the sample is restricted to adult children who are no longer in the home and for adult children who are 25 or over). Traditional notions of welfare cultures are ruled out of this setup, as are any explanations that focus on poverty preventing investment in children’s human capital.

2)      The authors find no evidence of the parents’ DI receipt affecting the adult child’s receipt of any other form of social assistance – which doesn’t fit the idea that there is a general reduction in the stigma of claiming benefits.

The authors conclude that the most likely mechanism for this result is that parents with hard-to-define disabilities who do eventually receive welfare receipt are better placed to educate their adult children in how to be successful in claiming DI. If the adult child has a verifiable disability, this could therefore be a welfare improving channel where their access to a required assistance programme is made more efficient. This is a very different context then from a traditional ‘family welfare culture’ story which is passed on through lower expectation and aspirations throughout childhood.

On Norway, work and welfare

It’s worth dwelling on a final point here – that the proportion of Norwegians claiming DI is relatively high in international terms and benefits are relatively generous. AT THE SAME TIME  the overall employment rate is higher than in the UK and the US.  In every age group from 20-24 year olds to 60-64 year olds, Norway has more jobs as a share of the population than the UK/US – overall, before the recession has taken root in 2007, 82% of Norwegian 25-64 year olds were in work, compared to 76% in both the UK and US. (If anything the difference is even starker in 2012, at 81.5% NO vs. 75.5% UK vs. 72.2% US; all statistics from the OECD Employment Database).

Just as strikingly, the employment rate among disabled people is higher in Norway than elsewhere (to the extent that we can measure this; see chart below).  And likewise, in a previous post, one of usshowed that countries with more generous welfare states tend to have a stronger ‘work ethic’ – or actually, are less sympathetic to benefit fraud – than countries with weaker welfare states, and this includes Norwegians being more strongly against fraud than Britons are.

Chart 2: Non-work by disability status in different welfare regimes

Figure on worklessness in different welfare regimes - see text 

Key: ‘Limiting l/t illness’ = those reporting a long-term illness that limits their day-to-day activities;
‘Healthy’=all others (note that there are uncertainties over the comparability of data on disability).
Source: Bambra, Work, Worklessness & the Political Economy of Health, based on van der Wel et al 2011


This raises a number of possibilities that might turn the dependency thesis on its head – it’s plausible that generous welfare systems actually help cement a commitment to work. In Norway it might just make more sense to work; working conditions are better, jobs are more secure, pay in low-skilled jobs is better, better childcare makes it easier for families to work, and people aren’t left clinging on to benefits for dear life in the face of strict conditions and Kafkaesque sanctioning regimes.  Just based on the patterns in this blog post, though, it’s also plausible that work is unusually important to Norwegian culture, and it’s this that makes it easier to sustain a generous welfare state.

What we do know, though, is that you can’t use this study of Norwegian disability insurance to argue for the welfare dependency hypothesis – as the authors themselves admit.

About Ben Baumberg Geiger

I am a Senior Lecturer in Sociology and Social Policy at the School of Social Policy, Sociology and Social Research (SSPSSR) at the University of Kent. I also helped set up the collaborative research blog Inequalities, where (after a long break) I am again blogging about inequality-related policy & research. I have a wide range of research interests, at the moment focusing on the role of social science, disability, inequality, deservingness, and the future of the benefits system, and I co-lead the Welfare at a (Social) Distance project (on the benefits system during Covid-19). You can find out more about me at
This entry was posted in Blog posts. Bookmark the permalink.

3 Responses to Killer evidence for intergenerational welfare dependency?

  1. Pingback: Contemporary Luddites, emerging England and intergenerational welfare dependency: Top 5 blogs you might have missed this week | British Politics and Policy at LSE

  2. terence says:

    That’s a great post – thank you!

    The paper came out yesterday or the day before as an NBER working paper and, while the authors may be careful in their claims in the text itself, from a quick read of the abstract I was left with the impression that the evidence *did* suggest welfare dependency of the classical type. And I owe you thanks for disabusing me of a belief that would have been very clearly mistaken.


  3. Pingback: When “Welfare-to-Work” doesn’t | The Community Chronicles

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.