Upfront, I should say that the debate about whether or not
something should be run by a private company or charity rather than the
government is not something that I find inherently interesting. I believe that
neither are the golden bullet that their supporters claim they are. More
important than who runs a service are questions of how that service is
resourced, how it’s run and how it’s evaluated. Last week Chris Grayling made amonumentally stupid comment in the House of Commons:
“The last Government was obsessed with pilots. Sometimes you just have to
believe in something and do it.” This
was in reference to his idea that the probation service should be privatised.
Now at the moment I have no idea whether or not this is a good idea but a
pilot* would be a good way of finding out. Mr Grayling is calling for policy by
ideology rather than policy by evidence. Whilst there is a necessary place for
ideology in government throwing out evidence is not the smartest idea. But I’ve
discussed that before so today instead I’m going to look at why payment by
results isn’t a magical cure-all.
To find an example
of payment by results being a comprehensive failure you have to look no further
than the last project Mr Grayling was in charge of, the Work Programme. The Work Programme paid providers for every
unemployed person that they got back into work for six months. Of course the
contracts for these schemes were not that simple because for making privatisation
work the devil is in the detail. A recent evaluation of the work programme
showed its most glaring flaw. The government was paying out millions of pounds
and were seeing fewer people go back to work than if there had been no program
at all. As flaws go, that one’s a big one. However, concentrating on the big
failure masks the little failures that still have a lot to teach us.
Firstly, the Work
Programme nurtured fraud on a large scale. A4E, one of the private
organisations that received some of the large contracts from the Work Programme,
was found to have been committing fraud in several of its services and to have
had a culture which fostered fraud. It was caught not because of government
checks but because of a whistleblower. If a scheme pays by results then it
encourages behaviour which will fulfil the checks, rather than necessarily
getting the results. Fraud is more likely to happen when you incentivise it.
There’s an interesting discussion on whether bonuses incentivise risky
behaviour and whether that was partly responsible for the economic crisis.
Payment by results may create a similar culture and need strong checking
systems to make sure the results are ones worth paying for.
Those who were given
priority in the work programme were not those that needed the most help butinstead those most likely to get a job.
As many of these were already perfectly capable of getting a job the work
programme took up their time from when they could have been seeking a job and
then got paid for “helping” them despite them already being job ready. This
artificially buoyed the performance of work programme contractors.
The Work Programme
also ended up damaging some of the most effective service providers. Although
I’ve not seen the evidence for it, many in the sector believed that small
charitable organisations were the most efficient at helping those with multiple
problems return to the job market. Additionally one of the stated aims of the
Work Programme was for more people to be helped by these small highly effective
organisations and for their funding to increase. However by many accounts the
opposite has happened. Due to the nature of the contracting process many of
these organisations saw a dramatic decrease in funding and have been forced to
close. This was not due to them being ineffective but rather because they had a
lack of referrals.
For the power of the
market to have full effect there needs to be true competition, innovation and
failure. A recent talk I went to featured John Kay from the FT talking about
what makes capitalism great. He used the example of the computing industry. No
single company had the all the successful ideas which ended up transforming the
sector and placing a computer in every home. Many of those that had successful
ideas still went bust as they failed to come up with the next idea. Public
service markets don’t work this way though. Provision comes from one of a
handful of very large companies, G4S, A4E, Serco etc. Government contracts work
against small organisations that would encourage innovation and against these
big companies failing if they can’t be the most efficient providers.
Small companies are
not the answer by themselves either though. The problem with using small
companies in payment by results schemes can be seen in TV license enforcement. In
TV license enforcement companies are paid for every household they find
illegally using a TV without a license. As anyone who has had contact with
these deeply unpleasant companies (or done any research on it) will know the
sector is rife with abuse, houses are illegally searched, people are pressured
into signing forms admitting wrongdoing when they have done and people are
frequently mislead as to their actual legal rights and duties. When misconduct
is found and punished rather than it being stamped out many of these small
companies dissolve and return to the same practices as a new company with
almost the exact same people.
The point I want to
make is not that payment by results is a bad idea but rather that it is very
difficult to get right. To neglect to do a pilot and get as much information as
possible to help design the best scheme that you can is a moronic show of
arrogance.
*Just to be clear
that this is not a partisan idea, a pilot of this very idea was set up by
Grayling’s predecessor Kenneth Clarke.
No comments:
Post a Comment