Some leisure activities are fixed in space or time, and hence cannot readily be altered by ICT.
For example, a Christmas-based family visit is fixed in time, although there may be spatial flexibility.
Location-specific outdoor activities, such as hiking in a particular area, are fixed in
space13, although there may be some temporal flexibility. Other leisure activities may not be tied to the intrinsic geography of a place, but to equipment or supplies that are stored there (a hobby like woodworking, falling in the bottom half of the figure, is a case in mind) – these also are less amenable to ICT alteration. Other activities (such as reading a book), falling in the upper right quadrant, are already both location and time independent, which also renders them less likely to be affected by ICT.

An arrow connects each of the activities in the lower left quadrant to its ICT-based counterpart.
Generally, these arrows point in a diagonal direction, right and up, implying greater flexibility in time and space. The arrowhead designates the frontier of the expanded choice set, with possible intermediate combinations in between. Some activities in Figure 1 are shown to have two arrows, indicating different impacts of ICT. Consider the case of attending a baseball or other spectator sports event. One possible ICT application allows one to hear on the radio, or see on TV, the action in real time, while not being there. Another possibility is to see the action in a time-independent mode via a recorded form. The three types of leisure activity, that of “being there”, “being there temporally but not physically” and “sharing the activity at a different time and place” constitute very different experiences, as noted by Katz and Dayan (1985).
For substitution to take place, the availability of an ICT-based alternative is a necessary, but not a sufficient condition. We believe that such availability is often not nearly as extensive as some would expect. For example, in a study of one week of activities conducted by each of 398 residents of Toronto, Canada in 2002-03 (comprising about 7,000 activities altogether), Doherty (2003) found that fully 80% of them were reported to have only one location at which they could occur.14 Temporal flexibility was higher: only about a quarter of the activities fell into the lowest range of a temporal flexibility indicator, while more than half fell into the highest range.
Conversely, as noted in the Introduction, many ICT-based activities do not have a location-based counterpart as a practical alternative, but simply would not have occurred otherwise (e.g., listening to a recorded performance on the radio). These types of activities fall into Category 2.

The degree of time- and location-independence of an activity may influence the choice set, but
does not determine choice. A number of studies have investigated factors influencing the
adoption of ICT activities such as telecommuting (Mokhtarian and Salomon, 1996), teleconferencing (Button and Maggi, 1994), and teleshopping (Salomon and Koppelman, 1988). In general, adoption is a function of the relative advantages and disadvantages of the ICT-based versus location-based alternatives, taking into account (as mentioned earlier) that the individual may value a number of factors beyond the surface ones. Mokhtarian and Salomon (2002) suggest a generic utility function for evaluating such alternatives, including variables such as the quality of the information obtained and the social/psychological content of the alternative. These variables often favor the location-based form of an activity over its ICT counterpart. Several dimensions relevant to the choice context and the characteristics of the alternatives are presented in Section 3.2 below.

3.1.2 Generation of New ICT Activities
ICTs offer opportunities for many new activities, such as playing games on a mobile phone. If
individuals spend more time on ICT-based activities, it stands to reason that they are spending
less time on non-ICT-based activities (with the exception noted below).15 To the extent that the foregone activities involved travel, this effect, like the preceding one, may also reduce travel.
Although the displacement may be immediate, as when an individual decides at a particular
moment to spend time on an ICT-based activity rather than some other activity, it can also occur over longer periods of time and more subconsciously than consciously. For example, when an individual experiences increased use of the Internet over time, each use specifically replaces another activity: either the non-ICT version of that activity (the kind of substitution described in the previous section), or some other activity altogether (a form of cross-substitution). But most likely that individual finds it more and more difficult to pinpoint exactly what activity has been “crowded out” by Internet use. As a result, the time displacement of other activities by ICT may be better captured by measuring longer-term trends in time use than by analyzing individual choices on particular occasions.
Available data indicate that Internet use and cell phone use have grown rapidly in recent years.
Given these increases, two questions arise: to what degree have ICT-based activities crowded out other activities (and to what degree will they do so in the future), and which activities get crowded out (and will in the future)?
The degree to which ICT-based activities crowd out other activities depends on the characteristics of ICT-based activities and the utility they provide relative to other activities. Characteristics that may tend to increase the utility of ICT-based activities include location independence, time independence, and fragmentability (see the dimensions discussed in Section 3.2 and summarized in Table 3). Utility will, of course, also depend on the technology. In general, as the technology improves, the utility of the activity will increase, and the potential for the ICT-based activity to crowd out other activities will increase.

However, the multitasking ability that comes with many ICT-based activities means that increased time devoted to these activities does not necessarily crowd out other activities. For example, when students talk to friends on their cell phones while walking across campus, they do not reduce time devoted to other activities,16 rather they do more with the time they have. The characteristics of location independence, time independence, and fragmentability also mean that ICT-based activities may get squeezed into the little blocks of time during the day that are too short or too inconvenient for other significant activities. In this case, ICT-based activities displace otherwise wasted time and also enable individuals to do more with the time they have.
Which activities will get displaced by increasing ICT use may vary considerably from individual to individual and from activity to activity. One might expect the activities most likely to be displaced over time to be those that offer rewards and satisfactions similar to those of the ICTbased activities that replace them. One might also hypothesize that the same kinds of activities that are more likely to be replaced by ICT versions of those activities, as described in the previous section, are also more likely to be displaced by increases in ICT-based activities more generally. But the characteristics of the displaced activity may not play as important a role in this case, given the unconscious nature of the displacement over time. Increased Internet use, for example, could be crowding out all kinds of activities, from doing jigsaw puzzles to attending concerts. Also, as the range of ICT-based activities increases over time, one type of ICT-based activity may crowd out another, as, for example, when Internet use leads to a reduction in TV watching.

3.1.3 ICT-enabled Reallocation of Time to Other Activities
The use of ICT may reduce the time and/or cost required to conduct activity X (or the travel
associated with X), with the saved time or money used (at least in part) to engage in activity Y.
For example, the travel time saved by telecommuting, or by videoconferencing instead of
traveling to a meeting, may be spent in part on leisure activities. The money saved by finding a
low-cost last-minute airfare on the Internet may be spent on other leisure trips and/or activities.
With respect to this category of impacts, relevant questions include: To what extent will timesaving ICT applications be adopted? How much savings will this mean? And how will the savings then be used?
As with the previous two categories, the extent to which time-saving ICT applications will be
adopted depends on the characteristics of the ICT-based activities and the utility they provide
relative to other activities. To the extent that the use of ICT in this context is a choice between
two forms of the same activity (e.g. commuting versus telecommuting to work), considerations
similar to those mentioned in Section 3.1.1 specifically for leisure activities apply. As usual,
whether or not the ICT-based alternative is chosen depends on the characteristics of each
alternative, the decision-maker, and the choice context.
The amount of time or money that is saved by a given ICT activity can depend on individual specific characteristics (e.g. one’s commute time, in the case of telecommuting) as well as on
technology (e.g. how effective an online “shopbot” is at identifying cost savings for a desired
item). Savings may not always be realized at all (or may be negligible), even in situations where they might be expected. For example, some studies show little or no cost savings achieved by Internet shopping (Brynjolfsson and Smith, 2000; Lal and Sarvary, 1999).

The time or money saved by ICT applications can be applied either to more ICT-based activities, or to non-ICT based activities, and to activities in any of the three basic categories. Thus, time saved by telecommuting might be used to work longer (mandatory), to cook more elaborate meals (maintenance), or to throw a Frisbee with the kids (leisure). The new mix of activities will again depend on individual-, activity- and alternative-specific variables. The effect on travel is ambiguous, depending on whether the new activities involve new travel or not. The evidence for telecommuting in particular is that the net impact is substitution, i.e. that the non-commute travel generation effect appears to be negligible and in any case far outweighed by the commute travel substitution effect (e.g., Mokhtarian, 1998).

3.1.4 ICT as Enabler/Facilitator/Modifier of Leisure Activities
Finally, the availability of ICT can facilitate activity generation and scheduling. For example,
mobile phones permit an impulsivity of activity engagement (spontaneous arrangement of
meetings; last-minute reservations) that was not previously possible (or at least not easy). By
providing readily-available information about an enormous variety of activity and travel
opportunities, the Internet facilitates making the arrangements for holiday and business trips, and may offer price bargains that allow more travel 17 to be consumed within a given budget. The result is at least a more flexible activity engagement, and potentially engagement in more out-of-

Continua >>>>>

13 Although this is true in a narrow sense, the ability of ICT to facilitate information-seeking and transactions (discussed further in Section 3.1.4) can broaden the choice set to include a larger class of “similar” locations. For example, instead of limiting one’s choice set of “great mountain climbing locations” to the Alps and the Rockies, browsing the Internet may expand it to include the Himalayas, the Andes, the Pamirs, the Karakoram, the Kunlun, and so on. The result is, in a broad sense, greater location independence.
14 Perhaps this figure was 95% a few years ago, and will be 70% in a few years. ICTs clearly are releasing some spatio-temporal constraints. Thus, we should not underplay this effect, but we should keep it in proper perspective.
15 Both the new ICT-based activities and the ones they displace could be non-leisure as well as leisure; for example, the ability to conduct business anytime, anyplace may crowd out leisure time.
16 Although it can be argued that they do, in fact, reduce time devoted to previously overlaid activities such as interaction with one’s surroundings and undirected contemplation. Different people will value this “lost” time differently.
17 Technically, to fall into this category the cost savings should result in choosing a more distant destination for a trip that was planned in any case. If the cost savings for one trip (or other purchase) is applied toward purchasing other trips or goods, it is an example of the third type of impact of ICT, discussed in the immediately preceding subsection.