The Value of Small Time Savings for Non-business Travel
A Daly, F Tsang, C Rohr, RAND Europe, UK
This paper looks at the evidence for valuations of small time savings as well as looking how these savings are treated in appraisal across a number of countries.
Travel time savings typically account for around 80% of quantified benefits of new transport infrastructure when formal cost-benefit analysis is undertaken and a substantial part of this benefit may be made up of small amounts of time. In most countries the value attributed to small time savings is assumed to be the same as values attributed to large travel time savings (per minute of savings). This assumption is often questioned. The argument is that no-one can get appreciable benefit from a saving of a few seconds but that many thousands of such savings, valued at the standard rate for value of time, may add up to large amounts of apparent benefit. This point of view is backed up by the findings of a number of Stated Choice (SC) experiments which have found negligible valuations for savings even up to five minutes.
Given the finding in the SC studies which underlie most valuations, the use of a uniform value requires justification and this paper will summarise the arguments that have been made in different countries. For example, has the main argument for uniform VOT been on economic theory or on the hypothetical nature of SC experiments? This paper will also summarise findings from SC work to give a nuanced description of the findings of most SC experiments that the size of the time saving considered affects the VOT significantly. Here we find that other issues, in particular the sign of the time saving, the trip length and the treatment of the status quo time and cost are intimately connected with the size effect and have to be considered also. We will review and summarise how values for travel time savings are specified across different countries, and whether special provision is made for small time differences, and how these vary across other dimensions, e.g. by mode, by distance, across income groups, etc.
We will also consider issues of forecasting that influence this issue. One relevant point is that when considering future transport schemes the notion of 'savings' is not relevant. What we are doing is not modelling the transition from one future context to another, where many people might experience small savings, but modelling the transitions from a base situation to alternative futures. In these transitions, people experience small or large changes, some are born, educated, retire or die and others change employment, home circumstances or location. The scenarios appraised therefore represent the differences between alternative futures. In this context it is simply not valid to identify the difference between one future and another as a time saving (or loss). The time taken for a journey has to be viewed as a whole and the fact that it might be a little different if a different decision had been taken is of no account to travellers. This reasoning would support the current attribution of a uniform value to time. The 'adding-up' argument has also been given weight in some countries, i.e. that it is not logical to value small differences brought about by part of a scheme at a lower value per minute than would be give if the scheme were considered as a whole.
Lastly, we will also consider the accuracy of our forecasting instruments. The identification of small differences in travel times for given journeys in two scenarios raises the question of whether those differences can be estimated accurately. Can network assignment models reliably predict travel times to within 1/2 minute? If not, then the reliability of the benefit calculations needs to be questioned more for schemes which offer small differences for large numbers of people than for schemes where the benefits are larger, though affecting smaller numbers.
Association for European Transport