Memorandum submitted by the Motor Cycle
Industry Association (MCI)
INTRODUCTION
1. The UK motorcycle industry turns over
in the region of £3 billion per annum and employs approximately
15,000 people
2. The Motor Cycle Industry Association
(MCI) was founded in 1910 and is the sole UK trade body representing
the non retail motorcycle industry. In this respect, the MCI represents
over 90% of the established UK industry, including manufacturers,
importers, accessory manufacturers, factors, related products
and services and some motorcycle dealers. .
3. The MCI is funded through membership
subscription, though the majority of its income comes from the
yearly Motorcycle Show at the NEC in Birmingham. This is organised
and managed by Motor Cycle Industry Exhibitions Ltd, a wholly
owned subsidiary of MCI and attracts between 150-200,000 visitors
per annum
4. The Motor Cycle Industry Association
(MCI) is pleased to contribute to the Public Accounts Committee
Enquiry into VED evasion.
5. The MCI would like to restrict its comments
to the important area of motorcycle VED evasion and submits specific
comments on the issue of how evasion rates are evaluated. The
MCI has strong concerns that the current methodology is producing
results that are both alarming and paint an inaccurate, indeed
sensational, picture of motorcycle VED evasion.
VED EVASION SURVEYS
6. The Department for Transport has conducted
regular roadside surveys over a number of years to provide data
concerning evasion of Vehicle Excise Duty (road-tax). By estimating
the number of vehicles evading VED a notional figure for lost
revenue can be calculated.
7. The VED-evasion methodology has remained
largely unchanged for decades, although traffic patterns (especially
motorcycle traffic patterns) have changed dramatically.
8. Observations are taken at a specified
group of 236 locations, equally dispersed between major and minor
roads in built-up and non-built-up areas, spread across the country
and selected to reflect general traffic patterns. There is an
unsubstantiated assumption that motorcycle traffic patterns must
match those of general traffic which may create a source of bias
in the samples due to the selection of survey sites.
9. The method used to estimate motorcycle
VED evasion assumes that any repeat sightings will be proportionate
to the vehicles total mileage. Again, there appears to be no evidence
to support this assumption for motorcycles as opposed to general
traffic. This assumption appears to have been last tested on keepers
of Heavy Goods Vehicles nearly a quarter of a century ago. It
seems highly unlikely that personal travel by motorcycle bears
much resemblance to patterns of commercial transport by HGV.
Source: Data extracted
from Transport Statistics Bulletins; Vehicle Excise Duty Evasion
(2004-06).
10. The number and proportion of valid observations
recorded for motorcycles each year is relatively low—in
June 2006, the survey yielded valid sightings of 9,309 individual
motorcycles, 0.9% of all vehicles in the survey during a month
when motorcycles accounted for 1.4% of traffic. The motorcycle
sample was just 0.8% of the licensed motorcycle stock in June
compared to the 3.5% sample of cars and 0.5% of the imputed total
motorcycle stock (assuming the VED-evasion estimated is reliable).
There appears to be a very strong association between the number
of valid sightings in the motorcycle sample on one hand and both
the size of, and confidence in, the estimate of motorcycle VED
evasion on the other—smaller numbers of motorcycles in the
sample have generated the highest estimates of evasion with the
poorest statistical confidence limits.
11. The (now) annual roadside survey takes
place in June. Motorcycling is very sensitive to seasonal factors
and June typically produces some of the highest levels of motorcycle
use, so it is all the more disappointing that the road side survey
design prevents a representative number of valid motorcycle observations
to be recorded. The DfT publication Compendium of Motorcycle
Statistics 2006 shows that motorcycling is concentrated into
the summer months, when there is about double the amount of motorcycle
travel seen in winter. The VED evasion survey does not account
for this seasonality and is consequently at risk of over-estimating
the general extent and cost of motorcycle evasion.
12. Until the 2004 survey, the DfT recognised
that the motorcycle sample size was too low to allow an estimate
of relative mileage to be used to inflate the final estimate of
motorcycle VED evasion. From 2004, the motorcycle sample size
effectively doubled (to its present disappointing level) and the
Department felt justified in applying the relative mileage methodology
to motorcycles to produce the extremely high VED-evasion estimates
seen in recent years. Whilst this is in itself a cause for contention,
DfT also chose to re-write the pre-2004 motorcycle VED-evasion
estimates (based on the then much smaller motorcycle samples of
less than 500 per year) as if these had suddenly gained the statistical
credibility the Department had previously stated they lacked.
The sample size has fallen by 10% since 2004, even though the
amount of June motorcycle traffic rose by 10% over the same period.
The motorcycle sample has fallen both as a total number and as
a proportion of all observations and as a proportion of motorcycles
in use.
13. Application of the relative mileage
methodology produces very wide ranges of statistical uncertainty
in the final results for motorcycle VED-evasion. This is manifest
in the extremely erratic estimates of motorcycle VED evasion levels
which now vary wildly from year to next (dotted black line in
the diagram above), in clear contrast to the trends seen in other
more directly measurable aspects of motorcycle use. The number
of motorcycles reported to have either current VED or SORN have
each been rising at rates that agree with other general indicators
of overall motorcycle activity (traffic estimates, number and
length of reported trips, collision rates, number of new and used
vehicle transactions, number of new driving licences issued for
motorcycles, etc.). Yet the rate at which motorcycle VED-evasion
is estimated to be rising is out of step and grossly out of scale
with these other indicators.
14. This is one of the most perplexing aspects
of the apparent motorcycle VED-evasion trends reported in recent
years; that evasion is apparently riding way ahead of growing
compliance rates. Vehicle Licensing Statistics 2006 reports the
number of motorcycles with current VED at the end of 2006 was
up by 1.5% and those with SORN declared up by 13%, significantly
ahead of the average for all vehicles.
15. That the number of motorcycles recorded
on the DVLA database either with current VED or SORN'd have both
been rising at the same time as VED-evasion would appear unlikely
in itself; that VED-evasion apparently rose by 47% in one year,
from an already highly-unlikely 471,000 evading motorcycles in
the year 2005 to 694,000 in 2006, simply beggars belief—especially
given that the number of evading motorcycles actually observed
on the road in the 2006 survey was less than 1% higher than the
previous year.
16. Data describing when an evading motorcycle
was last taxed/SORN'd, whether reported stolen, etc is not published
with the VED Evasion Survey, nor is there any data on follow-up
investigations regarding why evading motorcycles were in use.
Therefore it is difficult to suggest what mechanisms might make
some motorcycles more or less likely to be evading VED and from
there to suggest targeted solutions. However, we can say that
solutions which target the compliant vehicles (such as licensed-offender
style electronic tags) promise to be neither equitable nor effective.
We must also consider the costs of developing and administering
a higher compliance regime, who would bear them, whether they
would they result in a significant net improvement, etc.
17. Whilst it is clear that a single observed
use of a vehicle without VED counts as an act of evasion, it is
less clear whether total compliance with the licensing requirements
would yield significantly greater VED revenue. Faced with a total
compliance regime, some keepers would presumably choose not to
use their vehicle, while others would choose to continue riding
and to pay resulting in some extra VED revenue. However, total
compliance would also suggest compliance with the arrangements
for VED refunds, so it is by no means clear that any estimate
based solely on current apparent VED evasion can truly reflect
the extra revenue that is sensibly recoverable. Any meaningful
revenue loss estimate based on estimated instances of unlicensed
use, which (especially in the case of currently evading motorcycles)
may well be limited to very few months of the year, would have
to account for the net effects of refunds and compliance costs—the
current VED survey methodology does not address this.
IN CONCLUSION
18. All vehicle keepers have a social responsibility
and a duty in law to ensure their vehicles are compliant, motorcycle
users are no exception per se.
19. Any policy or method designed to combat
VED evasion should be proportionate to revenue loss and should
not penalise the majority of vehicle keepers who are compliant.
20. As a tool for informing policy makers
about targeting resources to reduce VED evasion in respect of
motorcycles, the current survey appears deeply flawed—there
is reason to believe the methods employed may grossly over-estimate
both the size of evading motorcycle stock and the consequent revenue
loss in represents.
| 
