Value of travel time for university students: a revealed/stated preference analysis

Abstract

Although the value of travel time (VTT) is extensively used in transport economics in order to conduct cost–benefit analysis and there have been numerous studies estimating this information for different countries, transport modes and type of users, there are very few estimates for the university students commuting from their hometown to the university-town. This article presents a VTT estimate for the students enrolled at the University of Trieste (Italy) based on sequential approach which: (1) collects revealed preference data in order to estimate a preliminary VTT; (2) generates C-efficient Bayesian designs pivoted around the characteristics of homogenous segments of the population and (3) administers an integrated revealed and stated preference questionnaire to a selected sample. The estimate obtained on the basis of the combined revealed and stated preference data ranges from 2.8 to 1.4€/h is significantly lower than the estimates obtained on the bases of the revealed preference data only, which ranges from 18 to 13€/h, and differs from the VTT for commuting to work reported in the literature by a factor of about 6.

Keywords:

• stated/revealed preferences
• travel time value
• university students
• commuters

Acknowledgements

We would like to thank Igor Sarman, Francesco Capodieci, Matteo Bino for their help in carrying out the project. The research is part of the UNIMOB project (Il Mobility Management per l'Università di Trieste), directed by Giovanni Longo, funded by the Commissariato del Governo nella Regione Friuli-Venezia Giulia, “Bando concorso contributi ambito ricerca universitaria anni 2007–2009”, also known as Fondo Trieste.

Notes

1. In 2011, there were 222 Italian towns hosting at least one university degree programme (CNVSU-MIUR 2011, p. 45).

2. Values are in 2011 prices.

3. “In contrast to orthogonal designs, so-called efficient designs do not merely try to minimise the correlation in the data for estimation purposes, but aim to result in the data that generate parameter estimates with as small as possible standard errors” (ChoiceMetrics 2011, p. 89).

4. The average fares by distance travelled to be paid to Ferrovie dello Stato Italiane, which is the most important Italian provider of train transport services, are published in the following website: http://www.fsitaliane.it

5. This would happen if the value of the marginal utility of the travel time is equal to 0.2, while the value of the marginal utility of income is equal to 0.000385; in this case, the ratio of the two parameters, representing the monetary value of travelling 1 h 40 times a month would be equal to 519, that is a VTT equal to 13€/h.

6. The a priori of the different WTP for living in different accommodation types have been proportioned to the new cost parameter estimated via the 35 SP interviews, keeping as a reference magnitude of the WTP, the values reported during the RP stage of the survey.

7. A comparison of the descriptive statistics of the two samples and the population is reported in the Appendix.

8. We constrained the parameter of the status quo specified in Equations (15) and (16) to be the same for the students living in TS and commuting.