Nowadays, if the sharing economy and its benefits are discussed, most will raise the matter of an increased socialisation and community growth, a greater focus on sustainability, and the improved utilitarian and cost aspects (Habib et al., 2017). However, a beneficial consequence of the sharing economy that is not discussed often enough is the actual social welfare opportunities it offers in many situations. Take as an example Uber, the ride-sharing service industry leader (MacMillan & Demos 2015). The company has been studied various times in the past in order to establish a link between its service and a decrease in motor vehicle accidents, specifically, fatalities. This relationship has recently also been successfully evaluated by Greenwood and Wattal, whose’ paper “Show Me the Way to Go Home: An Empirical Investigation of Ride-Sharing and Alcohol Related Motor Vehicle Fatalities” will be reviewed in this blog post.
What did they intend to analyse?
Overall, the researchers assessed what the exact impact is of ride-sharing services on alcohol related motor vehicle fatalities over time and the mechanics behind this specific impact. As a result, they split their research question into two hypotheses. The first was influenced by the platform theory, which states that consumers are willing to pay a premium for ride-sharing services, since customers are provided with the certainty of getting a car and knowing when this car will arrive, over the cost of searching for a random taxi and not even knowing whether they will find a taxi. Consequently, the first hypothesis was generated:
Implementation of a premium ride-sharing service will be associated with a negative and significant effect on the rate of alcohol related motor vehicle fatalities.
The second hypothesis followed the rational choice theory, where it is believed that if there is a high availability of ride-sharing vehicles, the inebriated individual might still consider their own car as a better option because the premium of the ride-sharing service is too high. In other words, the cost of hiring a driver might be too steep in their minds compared to the potential criminal cost of driving their own cars. Thus, the effect of a discounted service, which simultaneously increases the ease of use of transportation and decreases the gap between the cost of being punished for DUI and the cost of booking a driver, had to be assessed:
Implementation of a discount ride-sharing service will be associated with a negative and significant effect on the rate of alcohol related motor vehicle fatalities.
Which method was applied?
As a representation of the ride-sharing service, Greenwood and Wattal made use of information on Uber. More specifically, they analyzed Uber’s information between 2009 and 2014 in California and chose Uber X and Uber Black as representative services of a discount- (20%-30% under traditional taxi fares) and premium-service (20%-30% over traditional taxi fares), respectively. With regard to the dataset on motor vehicle fatalities, the researchers took advantage of the sources in the California Highway Patrol’s Statewide Integrated Traffic Report System (SWITRS). After combining both the Uber information with the SWITRS sources, a dataset of 12,420 observations spanning between January 2009 and September 2014 over 540 townships in the state of California, was able to be produced.
A difference in difference estimation was employed on the dataset, since it allowed for an imitation of an experimental design while using observational data, which led to the comparison of the number of fatalities changes after the application of the treatment over time.
What was found?
The results confirm that a discount ride-sharing service (e.g. Uber X) has a significant negative effect on the rate of alcohol related driving fatalities, while premium services (e.g. Uber Black) do not. Consequently, it can be determined that the cause for the decrease in DUI deaths lies in the combination of cost, availability, and ease of use, since consumers are not willing to pay a price premium. Moreover, it was found that the effect was significantly enhanced in larger cities and did not affect the overall fatalities rate. The latter finding disconfirms the common belief that by having introduced Uber and, therefore, more cars on the road, an increase in fatal accidents might have been caused.
These findings were also quantified, in order to increase the paper’s managerial relevance. It was determined that with only Uber X there was already a 3.6% to 5.6% decrease in alcohol related motor vehicle deaths per quarter in California. These percentages represent around 500 saved lives on an annual basis, which creates an additional public welfare of over $1.3 billion for Americans. Consequently, the paper truly confirms the social benefits ride-sharing services and the sharing economy in general generate.
These findings have implications for various professionals. First of all, it has direct implications for regulators and policy makers, who are currently assessing the legality of ride-sharing services. These results provide the necessary evidence of the sharing economy’s nontrivial effect, such as decreased mortality. This effect also impacts the second group of professionals, namely venture capitalists who can more easily be convinced by the idea of this social benefit and how it can be marketed. Finally, a specific group of professionals who gain a new strategic insight due to this paper’s specific results are restaurateurs, event planners, and nightlife managers. These professionals can use ride-sharing partnerships in order to promote themselves as safe environments after their clients leave their local’s inebriated. Furthermore, it is considered by many to be a sign of prestige to have a ride-sharing partnership, since it comes close to the traditional idea of a chauffeured service.
The paper poses a considerable strength when compared to past studies that analysed the link between ride-sharing services and their negative effect on DUI scenarios. Most of these studies were considered as invalid sources of information because they either involved the ride-sharing companies in the data analysis, their studies’ methodological rigor was invalid, or did not solve the presence of cofounding factors. In contrast, Greenwood and Wattal were able to solve these different issues in their paper. A great example of how they went above and beyond in order to validate their methodological rigor was through their several robustness checks, such as count models (e.g. OLS and QMLE), introduction of information of other ride-sharing providers, a coarsened exact match, a different data generation process, and a diagnosis of standard errors. And as if this would not be enough, the researchers covered most of their cofounding factors by performing a empirical extensions section that covered topics from the effect of local populations to a comparison of the alcohol related motor vehicle fatalities to the non-alcoholic ones.
A considerable weakness of the paper that was observed is its inability to account for random external causes for the accidents. While the reports of the SWITRS base cover what the cause of the accident was and what type of weather took place during the time of the accident, a lot of other factors were not able to be accounted for during the paper’s analysis. Furthermore, all analyses were not performed in a randomised manner, which decreased measurement validity and measurement reliability considerably.
Greenwood, B. N. and Wattal, S. (2017) “Show Me the Way to Go Home: An Empirical Investigation of Ride-Sharing and Alcohol Related Motor Vehicle Fatalities”, MIS Quarterly, 41(1), pp. 163–189.
Habibi, M.R., Davidson, A. and Laroche, M., 2017. What managers should know about the sharing economy. Business Horizons, 60(1), 113-121. Links to an external site.
MacMillan, D. and Demos, T., 2015. “Uber Eyes $50 Billion Valuation in New Funding,” The Wall Street Journal, May 9.