Third spaceor“infinite occasion”: Location choice and influencing factors of the new retail industry

With the development of technologies such as the Internet and artificial intelligence, the development of the new retail industry in urban areas has a differentiated choice for location. Using nearest neighbor index and kernel density analysis, this study took Starbucks and Luckin Coffee of Shanghai in 2018 as examples to examine the difference in spatial patterns between traditional retail and new retail. Spatial regression models were used to explore the differences in the factors affecting location choice between new retail and traditional retail. The main conclusions are that: 1) The location strategy of Luckin Coffee and Starbucks stores follows Christaller's central place model at the city level, but there are also differences between them. As a representative of the new retail industry, Luckin Coffee is more homogeneous in the urban circles. It does not place as much emphasis as Starbucks does on the central location of the city. 2) Overall, traditional retail and new retail have both similar and dissimilar characteristics on aggregation degree. The aggregation degree of Luckin Coffee is higher than Starbucks. In the central area of the city, traditional retail presents a single- central agglomerated feature, while new retail has a multi- centered dispersal feature. The aggregation degree of Luckin Coffee is lower than Starbucks. 3) There exists a major difference in the location of new retail and traditional retail—with a commercial center preference, the traditional retail Starbucks has a highly significant correlation with the density of shopping malls; but the most important factor affecting the location of Luckin Coffee is the density of office buildings. Rent has a negative impact, indicating that the new retail has a business office center preference.