A Regression Model of the Number of Taxicabs in U.S. Cities

One of the problems that face many cities in the country is being able to regulate the number of taxicabs that are operating in their areas. Not having enough or having too few can have adverse affects on the service, availability, and economic viability of the taxi business in their areas. It is the intent of this paper to take into account the taxicab population in 118 U.S. cites, and from this information generate a regression model for determining the appropriate size of the taxi business any comparable peer cities.

Deciding how many taxicabs to license in any jurisdiction is one of the most important decisions that the taxicab regulatory authorities of the area can make. Not having enough taxicabs creates lengthy waits and maybe even preventing customers from receiving service at all. Having an oversupply of taxicabs, on the other hand, can lead to improperly maintained taxicabs and potentially underpaid and poorly qualified employees. There are several methods that are being used to regulate the number of taxi licenses that are issued. The simplest of which is to just freeze the number of taxicabs that are in operation. Another approach is to require the taxicab companies to show that there is justifiable need for their services, and a third approach is to set a ratio between the number of taxicabs and some metric of population, taxi trip volumes, or other factor. This article uses a multiple regression model of the number of taxicabs in 118 U.S. cities and counties to develop a model that can be used by the appropriate regulators to assess the appropriate number of cabs for their jurisdiction.

In order to develop the regression model, the factors that affect taxicab demand need to be determined. Population is the most widely recognized factor, observing the number of taxicabs for 118 cities given; it is shown that there is a wide variation in the ratio of taxis to population. In order to arrive at a more accurate model, regulators often take additional factors of demand into account. Employment, other public transportation, business visitors, convention activity, and tourism are a few of the factors that can be taken into account to more closely estimate the levels of demand.

For the regression model in this article, the dependent variable was the number of taxicabs in 118 U.S. cities and counties with 100,000 or more population. The independent variables in the model were population, employment, vehicle ownership, transit use, airport passenger volume, and taxi fares. Using the available data for each of the independent variables, the correlation between each variable and number of taxicabs was examined. It was determined that three of the variables