Are Travel Time Savings Really Spent on More Working Time?

Are Travel Time Savings Really Spent on More Working Time?


S Torp Jespersen, Copenhagen Economics, DK; O Kveiborg,Technical University of Denmark, DK


We investigate how infrastructure induced travel time savings influence labor supply.


The decision about making transport infrastructure investments is very often based on the travel time savings that arises due to the new infrastructure. This has lead to numerous travel time valuation studies to find which value must be used in the socio-economic assessment. In the calculation of this value we need to consider two different values: The value to society and the value to the user. The latter is the value that determines the choices made by the user. In most studies the VoT is related to the wage rate. The theoretical difference between them is the income tax. The before tax wage rate is the value to society and the after tax wage the value to the user.

The changes in travel time valued using the VoT mentioned above implicitly assumes that all savings are spent on more work. However, this is most likely not the case. In order to find the true value of the travel time savings we need to find out how the time saved is spent.
We investigate the relation between travel time savings and the supply of labor. A standard economic model of consumer time consumption with leisure/labor and a general consumption good suggests that an increase in available time will be split between the two possible uses (leisure and labor). In the current context available time is defined as the time that is left when commuting has been carried out. Hence, any savings in commuting time will correspond to an increase in the "endowment" of time. If both leisure and the consumption goods are normal goods then an increase in income (the time endowment) will increase consumption. However, an increase in the consumer good can only happen if more time is spent on labor.

The qualitative analysis thus implies that travel time savings lead to an increase in working time. The model does in itself not tell us how the travel time savings are split between the two uses.

We use two different working time databases, which both include information about the working hours every year. The first database is based on self-reported working time and is a pseudo-panel since each respondent takes part in the panel for 1.5 years before a new person (with similar characteristics) take over. The second database is an administrative register which contains actual working time at the level of the individual employee reported by the employers. For a large share of the individuals in this second database they have a fixed working time according to their employment contract. It is this working time that is reported, even though actual working time may differ from the contractual working time.
We combine working time information with data on socio-economic characteristics from administrative registers. We have data on age, gender, education, address of work, home address, earnings, partner earnings, number of children, age of children, industry of work, and other background variables.

To investigate the impact of travel time savings on working time we have identified 7 infrastructure projects during the period 2001-2006. The information about working time covers the period from 2000 to 2007, which implies that we have information about working time before and after the changes in travel time. We compare the subsample of travelers that are likely to be influenced in their commute by the changed infrastructure with a control group with similar characteristics, but which is not likely to be influenced in their commute. We estimate a model on changing working time and control for a set of variables.
The estimations indicate a positive impact on working time. This is especially the case for women, which in fact start from a lower level of working time compared to men. In the paper we discuss our empirical results and suggest how our results can be used to adjust the currently used VoT for socioeconomic assessments of infrastructure projects.


Association for European Transport