The aim of changing different attribute levels in the stated preference data to explore the responsiveness of respondents on the choice between online shopping and in-store shopping, and the results will establish the policy implications. Marginal utility corresponding to the ratio of substitution between these attributes, for example, the valuation indicator of travel time is very important in transportation science (Louviere et al., 2000). How people trade-off the ordering time, shopping time as well as travel time and
Conditional logit Mixed logit Willingness to pay (WTP)
delivery time, when directly encounter the alternative-specific attributes in two shopping channels, under the obvious identifying experimental conditions. The valuation indicators are known as the willingness to pay (WTP), which is calculated by a ratio between two estimated coefficients in the case of linear additive utility function. The WTP is calculated as the average individual in the basic model [as shown in the equation from (5) to (8)] and the full model [as shown in the equation from (9) to (12)], which based on two methods, conditional logit model and mixed logit model, presented in Table 4.8.
Table 4.8: Willingness to pay of time and cost attributes
Basic model Coef. / (SE)
Full model Coef. / (SE)
Basic model Coef. / (SE)
Full model Coef. / (SE)
WTP for reducing ordering 153.47 154.37 166.17 128.28
time online (VND/min) (158.33) (130.98) (125.74) (102.40)
WTP for reducing shopping 40.66 -56.98 79.89 8.63
time in-store (VND/min) (146.43) (120.86) (113.37) (87.26)
WTP for earlier delivery (VND/day)
36,065***
(7895.45)
29,710***
(6597.94)
19,028***
(3050.43)
15,617***
(2527.81)
WTP for reducing time 414.17 -46.44 296.24 7.58
travelling to store (VND/min) (391.05) (413.13) (200.11) (215.25)
***: Significant at the 5% level
Regarding the WTP of conditional logit model, in the basic model, the WTP of one minute ordering time online as a ratio between estimated coefficient of ordering time and estimated coefficient of in-store purchase price, which has the value of 153.47 (VND/min).
This means that reducing one minute searching and placing the order over the internet generates the monetary value of VND 153.47. Meanwhile, one minute shopping time in- store produces the monetary value of VND 40.66. For total sample, the average time spend for searching and placing the order revealed to online shopping is 27 min (as shown in Table 4.4), which generates the monetary value of VND 4,143.69 (=27*153.47). The average
shopping time revealed to the bookstore is 50 min, so the corresponding monetary value is VND 2,033 (=50*40.66). In general, saving ordering time and time spent in store does not significantly improve welfare of shoppers.
The WTP of one day earlier delivery is calculated as the ratio between estimated coefficient of delivery time and estimated coefficient of in-store purchase price, which has the value of VND 36,065.41. Likewise, the WTP for reducing one minute travelling to store as a ratio between the coefficient of travel time and the coefficient of in-store purchase price, which has VND 414.17. In the revealed preference data, the respondents have to wait for delivery purchased products is 3 days on an average (as shown in Table 4.4). This waiting time would generates the monetary value of approximately VND 108,000 (=3*36,065). For the average travel time spend to go to store revealed 20 min, which produces the monetary value of VND 8,280 (=20*414.17).
In the basic model, those who are the online shoppers care much more about their value of time. Especially, they also considered the travel time as a waste of individual’s time resource. For example, the time spend for driving motorbike or car (in vehicle time), walk time, wait time and may other things, this can hardly do anything else. Instead of travelling to the store, the online shoppers are willing to pay more to save their time. A trip to a bookstore and back will spend 40 min (=20*2) of travel time and travel cost is VND 11,000 (as shown in Table 4.4), which generates the total monetary value of VND 38,560 [=(11,000 + 8,280)*2]. In this event, however, the online shoppers are willing to pay of VND 36,065 for one day delivery time of purchased products.
Regarding the full model, the WTP of one minute ordering time online is VND 154.37, the value does not differ significantly compared to the basic model. The WTP of one minute shopping time is VND 56.98. Note that the negative sign means those who are in-store shoppers, if the shopping time increases, then they will increasingly prefer. Although, saving ordering time and shopping time spent in store does not considerably enhance the welfare of shoppers. One day waiting for the delivery of purchased products has the monetary value of VND 29,710. Meanwhile, the WTP of one minute travelling to store is VND 46.44. A trip to a bookstore and back would generate the monetary value of VND 23,857.6 [(= 20*46.44 + 11,000)*2]. To save 40 minutes spend to go to a bookstore and come back which produces
the monetary value of VND 23,857.6, the online shoppers are willing to pay of VND 29,710 for one day delivery time.
In short, the WTP are calculated from the estimation results of conditional logit model which shows that, in the basic model, saving ordering time and time spent in store does not significantly improve welfare of shoppers. However, the the WTP of deilivery time affects siginificantly to the shoppers. Instead of travelling to store and come back (40 min), which produces the monetary value of VND 38,560, the online shoppers are willing to pay of VND 36,065 for one day delivery time. In the same manner, the full model indicates that instead of a trip to go a bookstore, the WTP of one day waiting for the delivery of purchased products is VND 29,710. This also implies that if the delivery time is delayed one day, then the purchase price of books over the internet will have to eliminate by VND 29,710 (if full model) or VND 36,065 (if basic model) to capture the in-store shoppers.
For the willingness to pay of time and cost attributes by mixed logit model is calculated similarly with the conditional logit model, which is a ratio between the two specific estimated coefficients, as previous mentioned. In the basic model, the WTP for reducing one minute for searching and placing the order over the internet is VND 166.17.
Meanwhile, the WTP for reducing one minute spend for finding and buying the items of books at a bookstore is VND 79.89. The average time spend for searching and placing the order revealed to online shopping is 27 min (as shown in Table 4.4), on an average, which generates the monetary value of VND 4,486.59 (=27*166.17), which is not different from the value calculated from the conditional logit model. The average shopping time going to the bookstore is 50 min, so the corresponding monetary value is VND 3,994.5 (=50*79.89). This value is almost double compared to the value of conditional logit model. However, saving ordering time and time spent in store does not significantly improve welfare of shoppers.
The WTP for reducing one day waiting for the delivery of purchased books is VND 19,028. Likewise, the WTP for reducing one minute travelling to store is VND 296.24. In the revealed preference data, the respondents have to wait for delivery purchased products is 3 days, on an average (as shown in Table 4.4). This waiting time, if eliminated, would generates the monetary value of VND 57,624.75 (=3*19,028.25). For the average travel time spend to go to store revealed 20 min, which produces the monetary value of VND 5,924.8 (=20*296.24). A trip to a bookstore and back will spend 40 min (=20*2) of travel time and
travel cost is VND 11,000 (as shown in Table 4.4), which generates the total monetary value of VND 33,849.6 [=(11,000 + 20*296.24)*2]. In this event, however, the online shoppers are willing to pay of VND 19,028.25 for one day delivery time of purchased products.
In the full model, the WTP of one minute ordering time online is VND 8.63, the value differs significantly compared to the basic model. The WTP of one minute shopping time is VND 128.38. Nevertheless, saving ordering time and shopping time does not enhace significantly the welfare of shoppers. One day waiting for the delivery of purchased products has the monetary value of VND 15,617.24. Meanwhile, the WTP of one minute travelling to store is VND 7.58. Instead of a trip to a bookstore and come back, for total sample, which generates the monetary value of VND 22,303 [=(11,000 + 20*7.58)*2], the online shoppers are willing to pay of VND 15,617 for one day of the delivery time.
In brief, the WTP are calculated from the estimation results of mixed logit model which changes considerably the valuation indicators, however trade-off between time and cost attributes and the welfare of shoppers stay the same. The noticeable findings show that, in the basic model, saving ordering time and time spent in store does not significantly improve welfare of shoppers. However, the the WTP for earlier deilivery appear to be considerable to the shoppers. Instead of travelling to store and come back, which produces the monetary value of VND 33,849, the online shoppers are willing to pay for VND 19,028 for one day earlier delivery time. In the same manner, the full model indicates that instead of a trip to go a bookstore, the WTP of one day waiting for the delivery of purchased products is VND 15,617. The comparison between the WTP of delivery time and WTP of travel time, shows a comparative advantage for online shopping.
In the previous study, Hsiao (2009) shown the WTP of travel time is 5.29US$/hour, while value of delivery time is $0.53/day. The monetary cost of a two-way trip to a bookstore amounted to USD 5.58 and average 5.48 days waiting for a purchased book that is equivalent a monetary value of USD 2.9. In addition, Schmid et al. (2016) also found that in Switzerland, the value of travel time saving (40 CHF/hour) is also high comparing to the value of delivery time saving (16 CHF/hour), showing a comparative advantage for the online channel.
60