Competitive Location Assessment – the MCI Approach

Abstract

The most important decision, which to a great extent determines the success of any retail outlet, is the choice of its location. In order to achieve the best possible capture of customers from the location we choose, our choice should be led by our knowledge about the store choice behaviour of customers in the trade area of our new outlet.Since Hotelling (1929) and Christaller (1933), economists have tried to derive the optimal number, size and placing of retail outlets by assuming that utility maximising individuals will shop at the closest store and tend to minimise the costs of shopping. Yet every day we observe that customers travel to very distant sites to shop just because of its particular features that differentiate it from other stores, so we may assume that an individual's utility is affected by the characteristics of the store where they shop.The main aim of this analysis is to identify the main determinants of store choice behaviour that cause these observed deviations from the predictions of simple models, with empirical evidence obtained by research about the store choice behaviour of hypermarket customers in Sarajevo, Bosnia. The survey amongst 240 households is conducted in Sarajevo, one of the fastest growing markets in transition and one of the least concentrated retail markets, which raises its attractiveness for the entrance of new international chains.The model employed for this purpose is the Retail Gravity Model, augmented by the inclusion of the MCI coefficient. The model incorporates both the attraction influences of stores' attributes and the deterring influence of the distance between store and customer on the customer's store choice behaviour, and the interaction of these two influences to estimate the probability of patronisation of particular stores by customers from particular areas.Since this analysis of the determinants of hypermarket store choice was conducted for the first time in Sarajevo, Abstract:In this paper, we have investigated the list of determinants that influence customers to create a short-term preference to a particular store. A survey among 240 households in Sarajevo was conducted in order to reveal their preferences. The results of the MCI model have identified several main stores' characteristics which have a key influence on customers' patronage decisions. The findings also suggest that customers, in making their patronage decisions, and hence making their preferences to a particular store, choose among a limited list of alternatives, which implies that competition among outlets is more space limited than had been expected.

Assessing the Utility of Retailer Based on Generalized Costs of End-Consumers

“…Businesses are paying a lot of attention to studying their customers, to increase their utility and comfort, minimize fatigue, and to attract choosing needed retailer. Generally, the utility function of end-consumers depends on the price (Nakaya, et al, 2000;Scott, 2017) and non-price factors (Birkin, et al, 2017; Mac-Innis, et al, 2014; Oruc and Tihi, 2012). At the same time, such methods do not fully reflect modern market conditions.…”

Attractiveness Modeling of Retail on Emotional Fatigue of Consumers

“…The attractiveness in the Multiplicative Interactive Choice (MCI) model is estimated using the multiplicative function defined on the values of the parameters of the perception of the store Nakanishi and Cooper (1974). Along with the numerical data, the model uses qualimetric indicators of service (Dummy variable) of service within the markets: the variety of assortment and quality of goods (Colomé and Serra 2000), the quality of service at the cash office (Colomé and Serra 2000;Oruc and Tihi 2012), any additional mass media advertising inside (Belch and Belch 2012), the tenant stores (Oruc and Tihi 2012); and beyond: access to the store by public transport (Colomé and Serra 2000), the convenience of working time (Colomé and Serra 2000), the way of getting to the store and back (Galkin et al 2018) and other. Values of them are defined by survey, questionnaire and other indirect methods.…”

Market Area Analysis for Retail and Service Locations with MCI

“…The Huff Model can be regarded as a theoretical base since Huff (1962) assumes size and transport costs as explanatory variables which can be tested by the MCI Model (Kubis and Hartmann, 2007;Suárez-Vega et al, 2015). But the number of additional influences tested is nearly unlimited and ranges from further store/location attributes like age or price level (Huff and McCallum, 2008;Tihi and Oruc, 2012) to the surrounding coupling and competition potential (Wieland, 2015a) to consumer-related subjective variables (Cliquet, 2013;González-Benito et al, 2000).…”

“…Thus, also market shares equal to zero, p ij = 0, what may occur, are invalid. A simple way to correct the raw data accepting a small bias is to increase the variable by a small constant (Kubis and Hartmann, 2007;Wieland, 2015a) and/or to aggregate the submarkets (Perales, 2002;Tihi and Oruc, 2012). Anyway, the raw data should be adjusted by removing singular instances and outliers to fulfil the LIFO/LOFI requirements mentioned above (Huff and McCallum, 2008;Wieland, 2015a).…”

“…A discussion of specific problems and opportunities regarding logarithmic transformations can be found in Silva and Tenreyro (2006) and Manning and Mullahy (2001). The violation of OLS assumptions and possible solutions have beed addressed in MCI studies several times (Nakanishi and Cooper, 1974;Kubis and Hartmann, 2007;Tihi and Oruc, 2012;Wieland, 2015a), but there are no corresponding extensions implemented in MCI yet. Other packages may be helpful for this analyses: since the implemented functions return "data.frame" (mci.transmat()) and "lm" objects (mci.fit()), respectively, this data can be processed e.g.…”