4. Submission from the Motor Cycle
Industry Association
Comments from the Motor Cycle Industry Association
(MCI) to the Minutes of Evidence taken at the Committee of Public
Accounts hearing on 23rd April 2008. (Winesses Dr Stephen Hickey
Mr Anthony Boucher and Mr Noel Shanahan)
GENERAL
MCI has been questioning the methodology of
the VED evasion survey since 2004, when the rate of motorcycle
VED evasion started to bear an increasingly lower resemblance
to known PTW registration and use patterns. Attempts to persuade
the DfT to review the VED survey methodology have been largely
ignored.
MCI noted the evidence given in response to
Committee's questions on 23 April.
MCI is given to wonder if the witnesses were
entirely open in their responses to the Committee questions. MCI
was intrigued by the repeated response that DfT were unaware of
the progress of the 2007 survey, or that there was any marked
differences in the data that was emerging.
SPECIFIC COMMENTS
Q3. Dr Hickey says a bigger sample size in 2007
(1.7M sightings, up from 1.3M in 2006) resulted in a more accurate
survey. However, this cannot explain the much lower estimate of
motorcycle VED evasion, because the number of motorcycle sightings
recorded in 2007 (9,166) were FEWER than the number in 2006 or
2005 (10,405 and 9,942 respectively). Also, 9,166 represents 0.5%
of the 1.7M total sightings, which is less than half of would
be expected from traffic counts at that time of year.
Q4. Dr Hickey says they used ANPR cameras to
capture the registration marks of 98% of the surveyed vehicles.
However, Appendix A of the VED Evasion Survey report from DfT
says the registration marks of ALL motorcycles were collected
manually (page 11, para 4). So, unless this is a grossly incorrect
statement by DfT then NONE of the massive reduction in motorcycle
evasion sightings can be down to the ability to double check everything
from a camera-image of the surveyed vehicle. The statement by
DfT also flatly contradicts previous suggestions that front-facing
ANPR accounted for a better motorcycle evasion estimate in 2007.
Q8. Mr Burr says NAO relies on National Statistics,
but surely a key task of an auditor is to check the whole process
from start to finish.
Q21. Mr Boucher says roadside clipboard accuracy
is checked at the roadside because there are two people involved
in recording "many" of the sightings but actually states
that one is reading the plate while another writes the number,
so there is no check on what is being read or written at the time,
merely two people splitting the chore.
Q26. Dr Hickey says that a roadside accuracy
check would need "... a parallel team of people checking
on a 100% basis". In fact, they would need a parallel check
of only a proportion of those readings. Given that the whole survey
was until recently based on manual readings, it would be quite
feasible to have more staff at the roadside to enable consistent
double-checking of the roadside readings now that these only account
for 2% of the whole survey.
Q45. Mr Boucher and Dr Hickey say of the University
of Southampton study into the VED Survey methodology that; it
looked at the middle of the estimation process, found the methodology
generally sound and made a few recommendation for the future.
But neither Mr Boucher nor Dr Hickey acknowledged that this was
only "part one" of a more wide-ranging study, for which
further funding was not forthcoming, thus preventing UoS from
looking at the effect on accuracy which many aspects of the VED
Evasion Survey in question.
Craig Carey-Clinch
Nich Brown
MCIA
6 May 2008
| 
