Supplementary memorandum submitted by
Transport 2000
At the Select Committee session on 23 November
2005 where we gave oral evidence, a number of points arose and
questions asked on which we considered it useful to do some further
work. What follows are some supplementary evidence and figures
that amplify points made in our original evidence and at the Committee
hearing.
1. INDEX OF
PASSENGER KM
AND EARNINGS
PER PASSENGER
KM 1989-2004
Source calculations from British Railways Board
Accounts/SRA, DfT and ONS websites.
Further calculations from the data reveal that
the average elasticity from 1989 to 1996 was around -1 and that
from 1999 onwards was -0.7. (Compared with -0.5 for the selection
of operators cited in our original submission). This implies that
there will a greater proportion of journey flows where the elasticity
is -1 or better. This means opportunities for operators to lower
the average earnings per passenger km and increase both passenger
numbers and their total revenue which also would result in modal
shift from in certain areas of the country very congested local
road networks. There are a number of factors which could have
led to the changes in fares elasticity before and after 1997 these
are for example;
(a) Introduction of fares regulation as the
consequence of the 1993 Railways act;
(b) The economic upturn;
(c) The consequence of abandoning the "fuel
duty escalator";
(d) The improvement in punctuality cited in our
previous additional submission to the Select Committee.
This additional research strengthens our original
case that there is a range of benefits (lower subsidies, reduced
congestion and pollution etc) from reducing overall earnings per
passenger kilometre. We would suggest to the Committee that they
ask operators to what extent they have examined fare elasticity
with the flows for which they have pricing control and how they
are utilising such information.
2. ELASTICITIES
MADE EASY
This section is a worked example of elasticities
in practice; we referred to this in our oral evidence.
PriceElasticity = %changeindemand/%changeinprice
This gives point elasticity which changes with where
you are on demand curve I tend to use average elasticity which
after some cute algebra reduces to:
EA = (Q2 ¸ 1)/(Q2+ Q1) (P2 ¸ P1)
where
Q1 = initial demand
Q2 = finaldemand
P1 = initialprice
P2 = finalprice
So based on Railway that is selling 100 £1
tickets a day:
| initial price |
final price | initial passengers
| final passengers | average elasticity
| change in fare box | change in pass numbers
|
| £1.00 | £0.75
| 100 | 150 | ¸1.40
| 13% | 50% |
| £1.00 | £0.50 | 100
| 220 | ¸1.13 | 10%
| 120% |
| £1.00 | £0.50 | 100
| 200 | ¸1.00 | 0%
| 100% |
| £1.00 | £0.75 | 100
| 120 | ¸0.64 | ¸10%
| 20% |
| £1.00 | £1.25 | 100
| 87 | ¸0.63 | 9%
| ¸13% |
| £1.00 | £0.80 | 100
| 111 | ¸0.47 | ¸11%
| 11% |
| £1.00 | £1.20 | 100
| 92 | ¸0.46 | 10%
| ¸8% |
| £1.00 | £1.50 | 100
| 95 | ¸0.13 | 43%
| ¸5% |
| £1.00 | £0.50 | 100
| 100 | 0.00 | ¸50%
| 0% |
| £1.00 | £0.75 | 100
| 100 | 0.00 | ¸25%
| 0% |
| £1.00 | £1.25 | 100
| 100 | 0.00 | 25%
| 0% |
| £1.00 | £1.50 | 100
| 100 | 0.00 | 50%
| 0% |
| £1.00 | £1.50 | 100
| 105 | 0.12 | 58%
| 5% |
| £1.00 | £1.25 | 100
| 110 | 0.43 | 38%
| 10% |
| £1.00 | £0.75 | 100
| 80 | 0.78 | ¸40%
| ¸20% |
| £1.00 | £0.50 | 100
| 50 | 1.00 | ¸75%
| ¸50% |
| £1.00 | £1.50 | 100
| 150 | 1.00 | 125%
| 50% |
| £1.00 | £1.25 | 100
| 150 | 1.80 | 88%
| 50% |
| |
| | | |
|
Worked example
So given that Northern Trains only receives 20% of its income
in fares then it's going on the basis of an elasticity of ¸0.64
that a fare reduction of 25% is going to cost the government an
increase of 2.5% in subsidy to achieve a 20% increase in passenger
numbers (ie value for money increase of 17%). So if the fares
went up by 25% then government subsidy would reduce by 2.25% and
value for money reduce by 11%!!!! If the elasticity is around
-0.5 then a 20% reduction in fares would increase the subsidy
by just under 5% but there would 6% increase in value for money—in
contrast a 20% increase in fares would reduce the subsidy by 5%
but reduce value for money by 3%.
The chances are that given the huge number of short journeys on
Northern Trains are short then the elasticities are probably nearer
or around -1 so it would not cost as much.
3. THE IMPACT
OF PUNCTUALITY
ON THE
DEMAND FOR
RAIL TRAVEL
Introduction
This section has been prepared in response to a question on the
impact of punctuality and factors other than fares on the demand
for rail travel raised by Mr Clive Efford. It is accepted that
a wide range of factors influence rail travel not just fares and
reliability factors but also frequencies and timings station cleanness
and security amongst many others. For instance, the Institute
of Transport Studies at Leeds University has calculated that there
is a strong correlation between economic growth and rail travel
so that for each 3% growth in GDP corresponds to a 2% growth in
rail travel. This paper does not seek to produce a "all singing
all dancing" model to predict rail travel, but instead focuses
on the impact of punctuality on the demand for rail travel and
how it impacts on passengers numbers and journey patterns. Comparable
data was readily available for the same sort of analysis that
was undertaken in the original paper submitted for evidence. However
in order to produce a quicker response the analysis below focuses
on each sector (i.e. Long distance, London and the South East
and Regional) rather than a selection of individual train operators.
Figure 1
Trends in train punctuality 2000—05
Source calculations from SRA
Figure 1 shows that punctuality has generally improved since 2001
and that it has strong seasonal trends with a low point in quarter
3 (September—December), though the difference between the
peaks and troughs is gradually diminishing. Further analysis suggests
that the moving average is improving by ½ % per quarter and
that if this continues it will be 2010 before a punctuality reaches
90%.
Methodology
Relevant data and statistics was taken from the SRA website for
the railway years ending 2001-05, with the year ending 2001 taken
as the base year for each set of data. The Microsoft Excel package
was used in each case to display the information graphically and
calculate correlations and the line of best fit so that the impact
of changes in punctuality performance could be calculated The
graphs for each set of analysis are shown in Annex 1. R2 (coefficient
of determination) is shown for each variable. The value of the
gradient for each graph represents 1% increase (negative = decrease)
for that variable for a 1% change in punctuality in the base year.
For example if in the base year the punctuality was 80% then the
gradient is the change for that variable for a 0.8% change in
punctuality. Results shown in italics are those derived where
the coefficient of determination is so small that it is considered
there is no relationship between that variable and punctuality.
LONG DISTANCE
OPERATORS
Table 1
Long Distance
| Punctuality %
| passenger km (billions | Revenue (£ billions)
| passenger journeys (billions) | average journey length (km)
| Earnings per journey |
2000-01
| 69.1 |
12.1 | 1.109 | 0.07
| 173 | £15.84 |
| | |
| | | |
| Long Distance | Punctuality %
| passenger km | Revenue
| passenger journeys | average journey length
| Earnings per journey |
| 2000-01 | 100 | 100
| 100 | 100 | 100
| 100 |
| 2001-02 | 100 | 107
| 109 | 106 | 101
| 96 |
| 2002-03 | 102 | 107
| 111 | 110 | 97
| 93 |
| 2003-04 | 106 | 110
| 117 | 116 | 95
| 92 |
| 2004-05 | 114 | 111
| 120 | 120 | 92
| 95 |
| R2 | | 0.60 |
0.74 | 0.81 | 0.55
| 0.09 |
| Gradient | | 0.56
| 1.13 | 1.23 | ¸0.83
| ¸0.17 |
| | |
| | |
|
Table 1a
| | passenger km (millions)
| Revenue (£ millions) | passenger journeys (millions)
| average journey length (km) | Earnings per journey
|
| Effect of 1% change in punctuality |
98.45 | 18.06 | 1.25
| ¸2.07 | ¸0.04 |
| as percentage of sector total 2003-04 | 0.01%
| 0.01% | 0.01% |
| |
| | |
| | |
Table 1a shows the results of calculations though the changes
are small compared with the sector totals the change in revenue
is equal to 15% of the total profits of the long distance sector
which in 2003-04 was just over £120 million. However in the
year 2003-04 an exceptional payment was made by the SRA Virgin
Cross Country of £58 million , if this taken into account
the total sector profit for the year falls to £62 million
and the effect of a 1% change in punctuality is equivalent to
29% of the total sector profits.
LONDON AND
THE SOUTH
EAST OPERATORS
Table 2
| London and South East | Punctuality %
| passenger km (billions | Revenue (£ billions)
| passenger journeys (billions) | average journey length (km)
| Earnings per journey |
| 2000-01 actual | 77.6
| 19.2 | 1.732 | 0.664
| 29 | £2.61 |
| | |
| | | |
| London and South East | Punctuality %
| passenger km | Revenue
| passenger journeys | average journey length
| Earnings per journey |
| 2000-01 | 100 | 100
| 100 | 100 | 100
| 100 |
| 2001-02 | 100 | 101
| 103 | 100 | 101
| 103 |
| 2002-03 | 102 | 103
| 102 | 102 | 101
| 100 |
| 2003-04 | 104 | 105
| 104 | 104 | 100
| 100 |
| 2004-05 | 109 | 110
| 108 | 113 | 98
| 96 |
| R2 | | 0.98 |
0.89 | .9942 | 0.75
| 0.79 |
| Gradient | | 1.05
| 0.71 | 1.38 | ¸0.30
| ¸0.61 |
| |
| | | |
|
Table 2a
| | passenger km (millions)
| Revenue (£ millions) | passenger journeys (millions)
| average journey length (km) | Earnings per journey
|
| Effect of 1% change in punctuality |
259.52 | 15.85 | 11.80
| ¸0.11 | ¸0.02 |
| as percentage of sector total 2003-04 | 0.01%
| 0.01% | 0.01% |
| |
| | | |
| | |
Table 2a shows the results of calculations though the changes
are small compared with the sector totals the change in revenue
is equal to 16% of the total profits of the long distance sector
which in 2003-04 was just under £97 million.
REGIONAL OPERATORS
Table 3
| Regional | Punctuality %
| passenger km (billions | Revenue (£ billions)
| passenger journeys (billions) | average journey length (km)
| Earnings per journey |
| 2001-01 actual | 81.7
| 6.9 | 0.572 | 0.223
| 31 | £2.57 |
| | |
| | | |
| | Punctuality % |
passenger km | Revenue
| passenger journeys | average journey length
| Earnings per journey |
| 2000-01 | 100 | 100
| 100 | 100 | 100
| 100 |
| 2001-02 | 97 | 101
| 90 | 100 | 102
| 91 |
| 2002-03 | 99 | 100
| 90 | 98 | 102
| 91 |
| 2003-04 | 101 | 109
| 96 | 108 | 101
| 89 |
| 2004-05 | 101 | 114
| 100 | 115 | 99
| 87 |
| R2 | | 0.50 |
0.60 | 0.56 | 0.57
| 0.03 |
| Gradient | | 2.38
| 2.04 | 2.82 | ¸0.44
| ¸0.44 |
| | | |
| | | |
Table 3a
| | passenger km (millions)
| Revenue (£ millions) | passenger journeys (millions)
| average journey length (km) | Earnings per journey
|
| Effect of 1% change in punctuality |
201.18 | 14.31 | 7.70
| ¸0.17 | ¸0.01 |
| as percentage of sector total 2003-04 | 0.004%
| 0.004% | 0.003% |
| |
| | | |
| | |
Table 3a shows the results of calculations though the changes
are small compared with the sector totals the change in revenue
is equal to 49% of the total profits of the long distance sector
which in 2003-04 was just over £29 million. However in the
year 2003-04 Gatwick Express as a result in the downturn in air
travel made a loss of £3.2 million, if Gatwick Express accounts
are removed from the sector accounts the sector profit rises to
£32.2 million and a 1% change in punctuality is equivalent
to 44.4% of total sector profits.
CONCLUSION
It is clear that for all sectors that punctuality performance
affects passenger numbers, though in each of the sectors the proportion
of journeys are relatively small. However the potential for changes
in revenue do represent significant proportions of overall profits
in each sector particularly for regional operators. The consequence
of this is that small decreases in punctuality performance can
affect profit margins and thus have the potential to lead to pressures
for fare increases and or further restrictions on "walk on"
fares such as "Savers" or additional subsidy payments
from the government, or a combination of all three of these. Conversely
small improvements in punctuality can increase profits substantially,
which implies that if current trends continue there will be less
pressure to increase fares and that as suggested in our previous
evidence to the Committee operators could look to either reducing
actual fares paid per mile, or lifting restrictions on "walk
on" fares or the DfT could seek to encourage modal shift
from the car and plane by using this "punctuality windfall"
to improve capacity.
Part of the analysis examined the effect of punctuality on average
journey length. The results show that in each of the sectors journey
lengths would decrease as a result of improved punctuality. However
parts of the various Route Utilisation Strategies started by the
SRA which in part are attempts to improve punctuality seek to
limit the availability of short journey opportunities. In other
words one possible outcome of some RUS will be contradictory,
our previous evidence to the Committee highlighted the high proportion
of journeys that are very short (ie 46% of rail journeys are less
than 20 miles). Has the Committee asked Network Rail if they have
taken this in consideration in the implementation of any or all
of the proposed RUS?
This evidence together with the evidence submitted previously
suggests that there is a relationship passenger travel and fare
level as well as punctuality. Has the Committee chair rightly
pointed out on the 23 November factors that influence the numbers
travel by train are numerous it is hoped that the evidence submitted
by TRANSPORT 2000 will unravel some of the complexity.
| 
