The Policy Exchange seminar on US welfare reform

A couple of weeks ago, I went to a seminar being run by the Policy Exchange think tank, where US welfare economist Professor Doug Besharov discussed the US welfare reforms that seem to be motivating the right and scaring the left in the UK at the moment. Interestingly, the main news to emerge was that neither side's views were grounded in the reality of the US welfare system. What follows are notes taken during his presentation, and those of fellow speakers Richard Johnson and James Bartholomew.

Doug's presentation

So what actually happened in the US?

• There was a 65% drop in unemployment claimant rates in the late 90s, following decades of rises.
• However, figures for benefit claimant activity in 2000 show that welfare-to-work activity was only being delivered to a tiny proportion of US benefit claimants.
• Additionally, unemployment fell across all 'barrier' groups, except for the 'never employed' (who correlated roughly with the 'multiple barriers' group)
• Doug felt that the increase in employment was primarily due to the economy. There was a huge rise in never-married mothers who went to work in the late 90s, due to strong labour demand including for less-skilled mcjobs.
• The mid-90s was also a period of major social change in the US. For reasons that are still not fully understood, US society became more conservative and law-abiding very rapidly, with drug use, crime and other indicators plummeting.

The keystones of US welfare reform were:

• Work First - requiring job search at the point of benefit entry immediately reduced entrants by 10-15%, an effect known as 'smoke-out' which is believed to remove people who are claiming fraudulently or have other sources of income.
• Sanctions - loss of benefits if behaviour-related rules are not followed.
• Time limits - this is entirely overstated. There are virtually no cases of claimants being thrown off benefit after five years, provided they're looking for work.
• Making work pay - In-work support doubled in the mid-90s and had a major impact on leaver rates.

What people did after leaving welfare:

• 40-50% of leavers went into full time work.
• Most of the remainder moved onto other benefits. Housing, Medicaid, food stamps and child support are completely unaffected by welfare reform. 50% of single mothers live with someone else, but the financial penalty on marriage of $6,000 / yr makes formalising the relationship unattractive. The impact on the average household is the loss of $1800 / yr, a small proportion of overall household income.

Conclusions from the US experience:

• Active labour market policies work.
• Job training schemes have not worked in the US, which raises interesting questions about similar schemes in the UK. In contrast, Doug felt that appropriate skills were key to people moving into work.
• We don't know who can or can't work - benefit reductions were roughly equivalent among all 'barrier' groups except those who had never worked. This raises some interesting questions about the use of creaming and parking.
• Many benefit recipients remain dependent until they retire, regardless of the support they receive.
• Making work pay works, kind of - Earning incentives (i.e. tax credits) encourage job entry but kill progression. The US incentives produced huge marginal tax rates on earnings above $9 / hr full time, leading to benefit leavers only taking on work equivalent to $9 an hour then sitting at that level.

James Bartholomew's presentation

Author of The Welfare State We're In and enemy of state benefits, education and healthcare James Bartholomew argued against the concept of benefits on the grounds that subsidising idleness inevitably led to increased unemployment, weakening moral fibre and lack of respect. Discussion of the points raised is beyond the scope of this article, as they were made without reference to empirical evidence, and they did not deal with the US experience or with any reform other than abolition.

Richard Johnson's presentation

A strong Serco contingent was present at the debate, which was otherwise surprisingly sparsely attended considering how close the Conservatives are to running the country. I didn't take as many notes during this part, mainly because I've been involved in everything Richard Johnson was talking about for long enough that most of what he said was unsurprising. Nevertheless it's worth reflecting that UK welfare reform may have something to show the US, rather than the other way round:

• The UK has its highest ever employment rate, its lowest unemployment rate in 30 years, and its lowest IB rate in 7 years. The employment figures are substantially better than those of the US.
• The UK already has an active labour market policy, primarily the New Deal. Richard has spoken to French welfare-to-work specialists who call Work First the Anglo-Saxon model. He highlighted what he thought was a major difference in the UK version of Work First - that it had been married to the European-style Social Model by Carey Oppenheim to produce a more caring, but still job-focused system.
• While the Work First model has worked up to a point, there are pockets of entrenched unemployment (geographical, ethnic etc.) that need something more to tackle them. Richard proposed a 'Work First Plus' approach to tackle these.

Some interesting discussion followed the presentation:

• MP Peter Lilley, who worked on the recent Policy Exchange report that compared different welfare systems, felt that only 'prime, working age males' should be counted in the employment figures, as the increase in employment under Labour was solely down to more women working instead of raising families. [This ties in with a pronouncement by David Cameron on the senselessness of increasing the employment figures by getting a parent to work while a non-parent looked after their children. Surely, removing jobs in childcare from the figures would prevent 'double counting' without pushing the assumption that a woman's place is in the home?]
• A DWP economist suggested that James' assertion that government statistics were obviously falsified or fiddled if they didn't agree with him was unfair to civil service statisticians. Richard Johnson also observed that the only recorded 'fiddling' was the Conservative government's targeting of jobcentre staff to push people onto IB 15 years ago as a way of reducing unemployment rates.
• The DWP economist questioned Doug's assertion that job training doesn't work alongside his assertion that the skills agenda was vital for tackling unemployment, primarily as a request for suggested solutions.
• Doug expressed a dislike of Work First Plus approaches - 'we've not learned how to fix broken pots' and should concentrate on improving education for the young to ensure that people don't enter long term unemployment in the first place.