International Choice Modelling Conference, International Choice Modelling Conference 2015

Font Size: 
The investment location choices of multinational enterprises: an application of the Latent Class Random Parameters model
Simona Rasciute

Last modified: 11 May 2015


The investment location choices of multinational enterprises: an application of the Latent Class Random Parameters model


This paper investigates the foreign direct investment (FDI) location decisions of multinational enterprises (MNEs) with an innovative application of the latent class random parameters (LCRP) model. Multinational enterprises face three major choices while undertaking foreign activities. The first choice is whether to produce at home and export, or whether to produce abroad. The second is that conditional on locating production abroad the firm has a choice among alternative locations of production. Finally, conditional on deciding where to locate, the firm has to decide on the scale of investment. Conceptualised this way, the first two decisions are discrete. The existing literature on FDI mainly analyses the first and third choices by assessing: why do some firms only serve the domestic market, while others also export or serve foreign markets through a subsidiary; and how much do firms invest? There is still a lack of detailed analysis of where firms choose to locate their investment. Of the studies that do attempt to investigate location choices of MNEs, most rely either on the Multinomial logit (MNL) or the Nested logit (NL) model. The MNL model, however, is subject to restrictive assumptions regarding the substitution patterns across alternative investment locations, while the NL model partially relaxes the independence from irrelevant alternatives (IIA) assumption in order to allow some substitution across alternative investment locations. Neither of these models allows for heterogeneity between investing firms, which is potentially very important for the choice of investment location as Nocke and Yeaple (2007) have demonstrated. Basile et al (2008) and Rasciute et al (2014) explicitly allow for the effects of unobserved investing firm heterogeneity on the investment location decision by using the mixed logit model to investigate investment location choices by MNEs.


This paper adds to this literature by extending the above analyses by explicitly allowing for two layers of unobserved heterogeneity in the investing firms’ decisions for the first time by using the LCRP model, which combines the latent class model with the random parameters model. The former assumes that investors are segmented into a predetermined number of heterogeneous classes, with the same parameters within each class but differing across classes (Greene, 2012). The allocation of investing firms into classes is based on observed as well as unobserved individual firm characteristics and it is probabilistic, so it is not known a priori which investing firm belongs to which class. The LCRP extends the latent class model by allowing for heterogeneity not only across classes but also within classes. Here parameters within each class have continuous variations and they are distinguished by different distributions.


A further innovation of the paper is that it employs a new multi-level data set, combining country-, industry- and firm-level data over a 16-year period to examine individual firm FDI location decisions in Europe. Country-level factors include such traditional FDI determinants as host country market size, taxation policy and transport costs, among others. Industry-level factors include dummy variables for scale-intensive, science-based, traditional and service sectors, as well as industry-level wages in the investment-receiving country. Firm-level characteristics capture investing firms’ size and profitability. By applying the LCRP model to multi-level data, this paper seeks to investigate how important the investing firm’s characteristics are in the choice of investment location and if alternative locations tend to attract different types of FDI, i.e. is the effect of location factors such as market size, transport cost and labour costs, among others, universal across different investors like it has been mainly found in the existing literature or it varies depending on investing firms characteristics? Furthermore, the LCRP model will further allow investigating if there is another layer of class-specific heterogeneity by imposing different distributions for random parameters within each class.


The initial results show that firms investing in different sectors and firms of different characteristics benefit from country-level factors to different degrees, i.e., the responsiveness of the probabilities of choices to invest in a particular country location to country-level variables differs both across sectors and across firms of different size and profitability. For example, firms investing in traditional sectors are less likely to be discouraged by higher host-country unemployment but more likely to be discouraged by higher wage rates, as compared to MNEs in non-traditional sectors such as science-based industries. These results support the fact that firms operating in science-based industries employ more skilled labour and higher wages may reflect skill premium. On the other hand, although unemployed people may lose their skills with time, firms operating in traditional sectors employ relatively more low-skilled workers and higher unemployment rates in some host countries may indicate higher availability of work force. Further initial results indicate that the larger is the host country, the more likely it is to be chosen by foreign investors; and the effect is stronger for larger investing firms, who need to exploit their economies of scale. On the other hand, more profitable firms are less likely to be discouraged from investing in more remote countries, as compared to less profitable firms, as they have more funds available to cover higher costs associated with carrying out activities abroad.




Basile R., Castellani D., Zanfei A. (2008). Location choices of multinational firms in Europe: The role of EU cohesion policy, Journal of International Economics 74: 328-340.


Greene, W. (2012). NLOGIT Version 5 Reference Guide. Econometric Software, Inc.


Nocke V., Yeaple S. (2007). Cross-border mergers and acquisitions vs. greenfield foreign direct investment: The role of firm heterogeneity, Journal of International Economics 72(2): 336-365.


Rasciute, S., Pentecost, E., Ferrett, B. (2014). Firm heterogeneity in modelling foreign direct investment location decisions, Applied Economics, 46(12): 1350-1360.

Conference registration is required in order to view papers.