Memorandum submitted by Fran Bennett (University
of Oxford) and Mike Brewer (Institute for Fiscal Studies)
A well-meaning desire to reduce complexities
in the benefit system can often be hindered by the difficulties
in defining "complexity" and in measuring whether it
A more productive approach would be to target
directly the outcomes which should improve if complexity is reduced.
These would include some outcomes already monitored
by DWP, such as higher take-up rates, lower levels of fraud and
error, lower administrative costs for DWP, a reduced total cost
to the Exchequer and a more acceptable distributional impact of
the benefit system.
But the impact of complexity on people receiving
benefits is not currently monitored by DWP. The most direct measure
would be the time and money claimants spend in claiming benefits,
maintaining a claim when circumstances change or benefits need
to be renewed, and also in stopping a claim where relevant. This
concept can be usefully thought of as the costs of compliance
with the benefits system.
We recommend both that the Committee consider
its terms of reference through the lens of the costs of compliance,
rather than solely benefit simplification, but also that the DWP
considering regularly monitoring claimants' compliance costs,
certainly for changes to benefits, but also for the existing structure.
Without this, there is a risk that policies will reduce the complexity
of benefits as seen by DWP staff without reducing the time and
effort spent by claimants.
1. We welcome the opportunity to provide
a short note to the Work and Pensions Select Committee's benefits
simplification inquiry. Whilst we appreciate the boundaries of
the Committee's remit, this note also covers tax credits. The
main purpose of the note is to suggest that the Committee could
usefully consider the concept of the "costs of compliance"
of benefits (and tax credits), and that this could provide a more
rounded and complete picture of the impact on claimants of the
design, mechanics and conditions of benefits/tax credits than
the narrower concept of benefit simplification (ie complexity).
Furthermore, given the difficulties in defining and measuring
complexity or simplification, there are considerable advantages
in targeting something which, if not straightforward to measure,
is at least one dimensional.
2. COSTS OF
2. There is an existing literature on the
compliance costs/burden of taxes (Evans, 2003)ie the costs
to the individual (or company) of complying with demands from
governments to pay taxes. These include the opportunity costs
of the time taken to fill in a tax return (for those for whom
this is relevant), the inconvenience of doing so, and any direct
financial costs incurred (such as employing an accountant).
3. We would argue that this concept of the
costs of compliance can also usefully be applied to those who
claim benefits/tax credits. It could be argued that people choose
to claim benefits/tax credits, whereas they are obliged to pay
taxes, and that therefore the two cannot be seen as analogous
in this way. However, we would argue that most benefits/tax credits
are essential in order to achieve an adequate income, obtain compensation
for disbenefits, meet additional costs (for example, of disability)
etc. And the existence of many benefits/tax credits also contributes
towards broader goals of economic growth, social cohesion etc.
to which this Government (like others) is committed.
4. Applicants and recipients of benefits/tax
credits incur coststime, financial and psychologicalin
meeting the requirements placed on them by social security/tax
credit law and statutory authorities. These requirements are not
only those associated with claiming benefits, but also those involved
in maintaining a claim and also in leaving benefit/tax credit
where relevant. The requirements could include, for example:
finding out about and applying for
taking part in any interviews required;
fulfilling reporting and other requirements;
receiving/obtaining the money;
dealing with any problems; and
any actions involved in leaving benefits/tax
Some costs may be incurred by members of their
families in addition.
5. These costs will obviously vary:
from benefit to benefit (means-tested
versus non-means-tested, but also those which require a medical
assessment versus those which do not etc);
from claimant to claimant (eg for
those for whom English is a second language compared with native
English speakers); and
with different circumstances (such
as time in the lifecycle etc).
6. This means that it would be impossible
to arrive at one global figure which could represent the "costs
of compliance" for each benefit/tax credit (though means-tested
benefits and income-related tax credits are likely to be the most
"costly"); and even to arrive at a range of estimates
which included such non-quantifiable elements as hassle, intrusion
and stigma etc. would be problematic.
7. But, however difficult the exercise,
we would argue that governments should be interested in these
costs, and in trying to measureand minimisethem.
For example, in line with the current Government's "rights
and responsibilities" agenda, the requirements of compliance
are increasing for some people (eg through penalties for non-attendance
at work-focused interviews, non-take-up of work-related activities
etc). And sometimes action to solve a problem can be accompanied
by increases in the costs of compliance for claimants; for example,
the recent 10-fold increase in the amount which can be earned
whilst on tax credits before income counts against entitlement
(from £2,500 to £25,000 additional income in one year),
which clearly reduces HMRC's administration and operational costs,
was accompanied by more stringent reporting requirements for changes
in other circumstances, which clearly increases claimants' compliance
costs. We believe that it is essential to monitor changes in requirements
such as these, in addition to any existing complexities of the
structure of benefits and claiming processes etc. It should be
noted that this kind of monitoring is likely to go beyond the
regular surveys of customer satisfaction which both the DWP and
HMRC already carry out (Sanderson et al, 2005; Herdan, 2006).
8. It could be argued that "compliance"
is too loaded a term, in that it is already in use to refer to
adherence (or not) to the rules governing benefits/tax credits
(the avoidance of fraud). An alternative phrase would be "transaction
costs"; this is a common term in economics, and refers to
the costs of locating information about opportunities for exchange
(transactions), of negotiating terms, and of enforcing the contract.
However, "transaction" tends to imply exchanges between
equals (as does much of the current government's language about
the "welfare contract") and that the costs are incurred
at a specific, discrete time; we prefer the term "costs of
compliance", despite the problems outlined above, as it implies
the inherent inequality of the position in which claimants of
benefits/tax credits usually feel themselves to be, and it also
recognises that the costs of complying with benefit and tax credits
can continue after the initial claim.
3. REASONS FOR
9. First, and most obviously, consideration
of the costs of compliance can help us to understand the reasons
for non-take-up of benefits/tax credits. Indeed, it could be argued
that this is the area in which such a concept is already well
used (see note 1). The literature on take-up already makes it
abundantly clear that complexity is not the only reason why some
people do not take up their entitlements. Costs such as stigma,
intrusion and hassle may be weighed in the balance against the
likely amount of benefit/tax credit to be obtained (Corden, 1999;
van Oorschot, 1991; Bunt et al, 2006)
10. Secondly, any cost and benefit analysis
of changes to benefits/tax credits that ignores compliance costs
is necessarily partial. Time spent by recipients fulfilling their
obligations is time that cannot be spent engaged in other activities.
Currently, government regulatory impact assessments (RIAs) estimate
the costs (including time) of new regulations to business, government
and the voluntary sector, but not to individuals. But an assessment
of the compliance costs to claimants of the operation of particular
benefits/tax credits, and of any changes to them, should form
an essential part of Government's consideration of whether to
introduce or amend social security provisions. The National Audit
Office has already recommended that the Government should aim
to minimise the compliance burden which its forms impose on citizens
11. Analysis of compliance costs can also
be used to develop a more rounded measure of the productivity
of the benefits system, following recommendations in the Atkinson
Review (2005) which suggested that ideally when examining productivity
the outcome for claimants should also be considered. (The same
could be argued for tax credits, especially when the compliance
costs of taxes are already under examination for employers by
HMRC.) There is considerable pressure within government to find
efficiency savings in the budgets for administering benefits/tax
credits (eg the current reductions in staff numbers and the closure
of local DWP and HMRC offices), and there is often a trade-off
between reducing compliance costs and reducing administration
costs. This trade-off will not be openly recognised if no effort
is made to assess and measure compliance costs.
12. We would therefore argue that the Committee
should consider its terms of reference through the lens of the
costs of compliance, rather than solely benefit simplification.
Reductions in complexity/benefit simplification may not always
be positive, holding other factors constant.
13. Instead, the Committee may instead wish
to consider the reasons that it is generally thought that a reduction
in complexity might be desirable. For example, a simpler benefits
system might lead to:
lower levels of fraud and error;
lower DWP(/HMRC) administrative costs;
a reduced total cost and a more acceptable
distributional impact of the benefit(/tax credits) system and,
we argue; and
lower costs of compliance for claimants.
14. We believe that it is these outcomes
as end goals which are crucial, rather than benefit simplification
for its own sake, and that one of these outcomes must be a reduction
in the costs of compliance for claimants. Currently, there could
be a policy bias within DWP because some of these outcomes are
explicitly measured and targeted by Government, but othersincluding
the costs of compliance for claimantsare not.
Review: Final Report (2005), Palgrave Macmillan.
Bunt, K et al (2006), Understanding
the Relationship Between the Barriers and Triggers to Claiming
Pension Credit, DWP Research Report 336, Leeds: CDS.
Corden, A (1999), "Claiming entitlements:
takeup of benefits", in J Ditch (ed), Introduction to
Social Security: Policies, Benefits and Poverty, Routledge:
Department for Work and Pensions (2006), Departmental
Report 2006, Cm 6829, HMSO.
Evans, C (2003), "Studying the Studies:
An Overview of Recent Research into Taxation Operating Costs",
e-Journal of Tax Research, Vol 1 No 1, pp 64-92.
Herdan, B (2006), The Customer Voice in Transforming
Public Services, Cabinet Office.
KPMG (2006), Administrative BurdensHMRC
Measurement Project: Tax, HMRC.
National Audit Office (2003), Difficult Forms:
How Government Agencies Interact with Citizens, HC 1145, Session
Sanderson, I et al (2005), Jobcentre
Plus National Customer Satisfaction Survey, DWP Research Report
282, Leeds: CDS.
Van Oorschot, W (1991), "Non-take-up of
social security benefits in Europe", Journal of European
Social Policy 1(1): 15-30.
3 April 2006
1 Though some of the literature on (non-)take-up attempts
to arrive at the "tipping point" where the amount of
benefit available outweighs the hassle, humiliation, effort etc.
involved in claiming it. Back