24 March 2015

A Good Source of Evidence

The House of Commons Work and Pensions Committee report "Benefit sanctions policy beyond the Oakley Review" is out today and has links in to lots of evidence - including our FoI request about Benefit Related Deaths - blog post here

23 March 2015

Sanctions help jobseekers into work of some sort

This may be the final post in this series that started here. You will recall that Esther McVey said there was evidence that the imposition of benefit sanctions helps claimants into work. Along the way we've seen mainly old, non-UK data used to support this. I asked the DWP if they had anything based on UK and/or recent data.

While waiting for that, I dug up a paper based on recent UK data that says that "sanctions are an ineffective tool for improving labour market performance." Along the way also I discovered another paper that says that "Benefit sanctions not only reduce unemployment durations but also reduce post-unemployment employment duration and earnings."

Today I got the response from DWP. No, they haven't  got anything based on more recent UK data. So I've thanked them and sent them details of what I found.

18 March 2015

Sanctions help jobseekers into poor quality work?

Some of the papers the DWP and I have been reading suggest that (imposed) sanctions help people into jobs while others say the contrary. I've found more work based on Swiss data that says "Benefitt sanctions not only reduce unemployment durations but also reduce post-unemployment employment duration and earnings."

This is quite logical. For a start, the data is from 1998-2003 when the Swiss unemployment rate was around 1.5-4.5% (See http://www.tradingeconomics.com/switzerland/unemployment-rate) so there may have been "scroungers" refusing to take available jobs. However, sanctioning them has pushed them into lower quality jobs. This is understandable too - jobseekers will have taken just about any job to alleviate the crisis but the sanctions actually harmed longer term prospects.  In more detail:

"The clear persistence of negative sanction effects on earnings up to two years after unemployment exit may be explained by lock-in into the accepted job or by faster return to unemployment. Once the individual has accepted a lower-quality-job, it may be difficult for him/her to catch up with the non-sanctioned people by quickly changing to a better job. Moreover, individuals who accept a worse paid job are more likely to leave this job and return to unemployment. Both lines of reasoning explain why sanctions lead to a reduction in post unemployment earnings."

When you study the DWP literature, it is all about getting people into work as soon as possible. The idea that being a more selective jobseeker could be a better long-term approach doesn't seem to have occurred to them.

Related posts here and here and now here

15 March 2015

Sanctions help jobseekers into work? (Contrary evidence)

I wrote recently about the debate over the effectiveness of sanction as a tool to get the unemployed to try harder and get jobs. The DWP had come up with a load of evidence that it is effective although that evidence is non-UK and old enough to be from a different employment climate. I've asked for more recent research based on UK data but in the meantime I've found some myself. It comes from Howard Reed of Landman Economics and is published by Oxfam.

HOW EFFECTIVE ARE BENEFIT SANCTIONS? An investigation into the effectiveness of the post-2012 sanctions regime for Jobseeker's Allowance claimants uses a technique similar to that used Boockmann to analyse data from Germany during 2006-7. Sanction rates vary from place to place - as I have shown myself - so this should be reflected in re-employment rates. Boockmann used this to show that sanctions helped people into work but Reed found "The results show no evidence that Jobcentre Plus districts with higher rates of sanctions between October 2012 and June 2014 experienced greater decreases in unemployment or increases in employment than districts where sanctions were used less often" and goes on to say "Overall, this report finds that sanctions are an ineffective tool for improving labour market performance."

A slight weakness in this paper is that it appears to accept the geographic variation in sanction rates as essentially random. There's some work by David Webster showing that sanction rates tend to be higher in areas of high unemployment. Perhaps staff are keen to encourage jobseekers to try harder in jobs deserts? Somewhat futile!

A further related post is here

10 March 2015

Sanctions help jobseekers into work?

Benefit sanctions are supposed to give jobseekers an incentive to take necessary action to get in to work. Agencies such as Citizens Advice say they can be counter-productive when actually imposed "Claimants are distracted from job-hunting as they focus on putting food on the table and keeping a roof over their head." In contrast, on 4 February 2015, Rt Hon Esther McVey MP, Minister of State for Employment told the Work and Pensions Committee "... and there is further research that, should somebody have been sanctioned, it helps them into work afterwards."

So I asked for the evidence and (unusually for the DWP) they sent it!

"The minister referred to a range of international evidence regarding the impacts on
employment following a benefit sanction. Recent studies from Denmark, Germany,
Netherlands and Switzerland show an increased employment uptake for recipients of
unemployment insurance or welfare benefits after a benefit sanction has been imposed"

See below for the research they've relied on together with some other data I thought was relevant:

Source Location Dates Typical SanctionUnemployment then/January 2015
van den Berg G, van der Klaauw B, van Ours J (2004) Punitive sanctions and the transition
rate from welfare to work. J Labor Econ 22:211–241
Holland 2003-2005 20% 1-2 months6%/7.2%
Abbring J, van den Berg G, van Ours J (2005) The effect of unemployment insurance
sanctions on the transition rate from unemployment to employment. Econ J 115:602–630
Holland 1992-1998(?) 5% 4 weeks?/7.2%
The Effect of Punitive Sanctions on the Transition rate from Welfare to Work Qureshi Denmark 2007-8 33% 3 weeks2.2%/4.6%
van der Klaauw B, van Ours J (2013) Carrot and stick: how re-employment bonuses and
benefit sanctions affect exit rates from welfare. J Appl Econometrics 28(2):275–298
Rotterdam, Holland 2000-2003 10%5.3% */7.2%
Svarer M (2011) The effect of sanctions on the exit rate from unemployment: evidence from
Denmark. Economica 78(312):751–778
Denmark 2003-2005 2-3 days or 3 weeks 15%5.7%/4.6%
The Effect of Benefit Sanctions on the Duration of Unemployment Rafael Lalive Switzerland 1997-8 3 weeks 100%4.5%/3.5%
Boockmann et al (2014) Intensifying the use of benefit sanctions: an effective tool to increase
employment? IZA Journal of Labor Policy 3:21
Germany  2006-7 10%9.5%/5%
Data source Click on the "Jobless figure for the country of interest. A * indicates incomplete data.

We could argue about the definition of"recent" but let's not other than to observe that the data for this research mostly pre-dates the general rise in unemployment during 2008.

During this period it was employers who had a hard time in the job market so lighting a fire under lazy "jobseekers" might well have persuaded them to take jobs that would otherwise have remained vacant.

At times of high unemployment it's still possible that some sanctioned jobseekers will compete very hard and get jobs that the non-sanctioned also tried for but total unemployment will be unaffected.

What would be useful is some research from a time of high unemployment, preferably from Britain and I've asked DWP if they have any However, the study in Germany at a time of 9.5% unemployment (Boockman) is close but it makes the usual mistake of  looking at higher employment of a small group subjected to a particular treatment while ignoring the displacement of their untreated competitors.

So, strictly speaking, DWP have answered the question. Things we don't know include:
  • How much evidence to the contrary is out there
  • If the DWP is aware of it
  • If there is more relevant UK evidence from a time of high unemployment


Severity of Sanctions 


This is another important factor. Most of the sanctions in the table above are less severe than current UK regime. In the UK, sanctions are for a minimum of 4 weeks and can go up to 3 years. Nominally they are 100% benefit cuts although the little advertised and restrictive hardship payments system can reduce them to to 40%. Form this is could be argued that the UK is more coercive leading to better results. There's two main objections to this logic:

  • The usual one about the futility of coercing people into non-existent jobs
  • Whether the jobseeker is close to destitution or not
To expand on the second point, some jobseekers have sufficient savings for a sanction not to be an immediate distraction from jobseeking while still unpleasant and "motivating". It's also plausible that for some claimants, it does (as mentioned above) render them "distracted from job-hunting as they focus on putting food on the table and keeping a roof over their head"

So, the research that's really needed should look at the effects of sanctions on near-destitute jobseekers. One way of detecting this is to look at the fate of sanctioned jobseekers  with very low savings. The DWP should have the savings data but so far as I know, they haven't attempted a correlation with fate.

Perhaps an FOI request for another day?

Part 2: Sanctions help jobseekers into work? (Contrary evidence)