Methodolgy
Methodology
This appendix explains in more detail the methodology used in this report to analyse the potential impact of a range of policies targeted at financially incentivising or disincentivising transport mode choices.
Qualitative research
We conducted a literature review of existing research from across policy, academia and the private sector, and analysed publicly available datasets. We carried out 14 interviews with policymakers at local, regional and national levels, within and outside London. The interviews were semi-structured and lasted from 30 to 60 minutes. Topic guides were tailored to the field of expertise of each individual and to address knowledge gaps identified during the literature review. The qualitative research was also used to identify policies to be tested in the financial modelling.
Financial modelling
Our quantitative analysis set out to model the impact of selected financial measures on the travel costs of a range of representative personas, explained in appendix 2. As such, the modelling was able to reveal potential “winners” and “losers” under these policies, as well as illustrate the scale of financial impact that these policies could have on the costs of travelling in London. The model does not tell us the modal shift implications of the measures. It was not possible to test every kind of policy using the model, but we have evaluated these elsewhere in the report using the qualitative research (e.g. mobility credits).
Modelled personas
The personas were created based on TfL’s Transport Classification of Londoners. We determined their residence by using the borough with the highest concentration of relevant residents. We also used the segment summary profile (including their life stage and preferred modes of transport) to determine their age, whether they have children, and the age of any children. We also determined the annual household income for the personas. Based on this information, we estimated the price of their cars and their VED bands. The price and age of the cars were then used to calculate whether the personas would need to pay the ULEZ charge when driving in London.
Modelled journeys
For each persona, we modelled the cost associated with a specific journey to understand the policy impacts on a range of journey purposes. These included trips to visit family, to undertake caring responsibilities, for evening leisure activities, to commute to work or university, and “trip chaining” journeys that go to multiple destinations. The journeys modelled also included a range of areas of London, with trips entirely in outer London or inner London, radial trips from outer London into inner London and a trip made to a destination outside London.
For each journey, we decided on the following:
- Starting point;
- Destination;
- Distance of journey;
- Whether the journey is possible via tube or bus;
- Whether the journey is cyclable.
To determine if a trip is cyclable, we used TfL’s 2016 cycling potential criteria and considered that all trips below eight kilometres (approximately five miles) were cyclable. 21
Costs of driving
The model calculated the cost of driving for each given journey. This cost incorporates both the use costs of undertaking that journey and the overall availability costs of owning the car.
Use costs of driving
The following variables were used to calculate the use costs of driving:
Fuel price and fuel duty
Fuel price is set at £1.47 per litre, correct as of 26 April 2023.
Fuel duty is a variable in the model, but as a default was set at the current rate of £0.53 per litre. Fuel duty forms part of the above price of fuel.
Electricity
For EVs, the model assumes an energy efficiency of four mile per kWh based on a Volkswagen ID.3.
Electricity price is calculated using a rate of £0.33 per kWh, correct as of 26 April 2023, assuming the vehicle is charged at home.
Congestion Charge
Congestion Charge is a variable in the model, but as a default was set at the current rate of £15.00 per trip in the zone.
ULEZ
ULEZ charge is a variable in the model, but as a default was set at the current rate of £12.50 per trip for eligible cars. It was assumed that the proposed ULEZ expansion has been implemented.
Destination parking
Cost of destination parking varies by persona as each borough manages its parking tariffs differently. For each persona, the parking cost at the destination was estimated assuming the time they would stay at their destination and searching the relevant borough’s website to confirm the tariff.
Road user charging
We assumed that the introduction of smart road user charging will need to generate at least a similar revenue to the charges it replaces (i.e. ULEZ and the central London Congestion Charge). ULEZ expansion is assumed to generate between £100 million and £300 million in the first year and be equivalent to zero after four years. The revenue generated by the Congestion Charge in 2022/23 was £84.4 million. In 2021, the vehicle miles in London amounted to 14 billion (excluding buses, LGVs, HGVs, motorcycles and pedal cycles).
The minimum price per mile was determined by dividing the revenue generated by the ULEZ (expectation of £300 million a year) and the Congestion Charge by the vehicle miles. This equates to £0.03 per mile.
This type of charging aims to encourage people to drive less, but also to drive cleaner vehicles. Using the existing VED, the charge per mile was increased by £0.10 for each VED band (so B is £0.13 per mile, C is £0.23 up to M at £1.23).
Availability costs of driving
The following variables were used to calculate the availability costs of driving on an annual basis:
Ownership costs
For each persona, we attributed a purchase cost for their vehicle ranging from £4,000 to £50,000. We also assumed the number of years that the persona had owned the vehicle.
If the vehicle was purchased more than four years ago, then the model assumes that the owner is no longer paying towards ownership costs.
For vehicles purchased less than four years ago, the model assumes that they are paying for the car on finance using a personal loan over four years with an APR of 12 per cent. 22 23
Insurance
The insurance cost was calculated at £470 per year based on the average cost of UK comprehensive car insurance. 24
MOT
The MOT cost was assumed to be the maximum price of £54.85, as set by the government. This cost applies to vehicles three years or older.
VED
Registration year VED: Each car was attributed to a VED “band” that was used to calculate its first year VED. This was a variable in the model, but as a default was set according to current rates. 25
Subsequent years VED: This was a variable in the model, but as a default was set to the current rate of £180 per year for petrol or diesel cars, and £0 per year for EVs.
Vehicles owned for less than five years and that cost over £40,000 are subject to an additional £390 of VED per year.
Norwegian-style VED
In policy paper 7, we tested a “Norwegian-style” model of VED for the first year of new car ownership. The values were modelled using only the CO2 tax, although in Norway they would also consider NOx and weight in determining the overall tax rate. Rates were taken from the 2023 tax rates on the Norwegian government website.
The CO2 tax is stepped, with rates calculated per g/km at increasing amounts. In the model, VED Bands A-C were set to zero, to account for the tax rebate that is applied in Norway up to 82g CO2/km.
In each VED band thereafter, the tax was calculated assuming the cars’ CO2 emissions were at the upper limit of the VED band.
An exchange rate of 1 NOK = 0.078 GBP was used, correct as of 13 March 2023.
Residential parking permit
Residential parking permit costs vary by borough. The model used the VED band of the car and the borough of residence to calculate the annual cost of a residential parking permit for personas who don’t have access to off-street parking.
To convert these annual availability costs into a cost per trip, the average number of miles driven each year by drivers in London (11,283.1 miles) was used to calculate an average cost per mile for each persona. 26 This was then multiplied by the distance of the journeys being modelled.
Costs of public transport
Public transport fares were variables in the model, but as a default they were set at current rates, correct as of 26 April 2023.
Journeys modelled were either taken by bus or tube and either during peak or off-peak hours. The model also considered current concessionary schemes and child rates.
Costs of cycling
To calculate the costs of cycling, we compared three cost options: Santander cycle hire, personal bike ownership, and personal bike ownership if purchased using the cycle to work scheme.
Santander cycles
The model compared three ways of paying for Santander cycle hire: PAYR, monthly subscription and annual subscription.
PAYR was set at £1.65 for each 30 minutes. 27 An assumed speed of six minutes per mile was used to calculate the multiplier applied to the PAYR rate.
The monthly subscription was set at £20 per month. The average distance cycled per year by cyclists in the UK (1,087 miles) was used to calculate an average cost per mile. 28 This was then multiplied by the journey length.
The annual subscription was set at £120 per year. 27 This was also converted into an average cost per mile and multiplied by the journey length. The model also incorporated discounts for the cycle to work scheme, NHS staff, students and recipients of ULEZ scrappage grants. Discounts were also used as potential variables but were set at current rates.
Personal bike ownership
For each modelled persona, we calculated the cost of purchasing the necessary bikes. The cost of owning a bike was treated similarly to the costs of owning a car, with costs split over four years and then written off after that.
It was assumed that servicing each bike would cost £50 per year. 30
The model also accounted for the cycle hangar cost in their borough of residence if they didn’t have space to store their bikes at home.
The total annual cost was then divided by the average distance cycled annually (1,087 miles) to calculate a cost per mile. 28 This was then multiplied by the journey length.
Cycle to work scheme
For personas eligible for the cycle to work scheme, we also looked at the impact using the scheme would have on their costs for cycling.
This was done by applying a discount of 35 per cent for personas earning £50,000 per year or more, and a discount of 25 per cent for personas earning less than this. This discount was only applied to adult bikes.
The discounted ownership costs were then combined with servicing and parking costs and converted into a cost per mile as outlined above.