The effect of host country Internet infrastructure on foreign expansion of Korean MNCs

Abstract

This study examines the effect of host country Internet infrastructure on a multinational corporation (MNC) foreign expansion. Using Heckman’s selection model on a sample of 2589 subsidiaries of 487 Korean MNCs between 1990 and 2011, we find that host country Internet infrastructure is important in MNC expansion decisions. In addition, we find that a well-developed Internet infrastructure within a host country leads to more investments from MNCs producing consumer over industrial goods and is more attractive to domestic market followers than market leaders. We find that the host country’s Internet infrastructure is important for an MNC foreign expansion decision, suggesting that efficient communication within an MNC is critical in coordinating globalized MNC subsidiary operations.

Keywords:

• Consumer goods versus industrial goods industries
• Internet infrastructure
• market leader versus market follower
• MNC expansion

Notes

1. Twenty-five host countries are Australia, Bangladesh, Brazil, Canada, China, France, Germany, India, Indonesia, Japan, Malaysia, Mexico, Netherlands, Philippines, Poland, Portugal, Russia, Singapore, South Africa, Sri Lanka, Thailand, Turkey, United Kingdom, the United States, and Vietnam.

2. For ‘controlling for self-selection, the Heckman-based approach has one major advantage: the self-selection parameter that gets carried into the second stage serves to ensure a quasi-simultaneity in estimation that strengthens the method’s ability to address self-selection bias’ (Martin 2013, 33).

3. This simulation approach estimates simulated 95% confidence intervals for the probability of expansion by randomly graphing a thousand normally distributed values of the coefficients on constitutive variables. We used this simulation approach because of the nonlinear nature of our models, logit models. When an interaction term is present, the confidence interval surrounding the estimated effect of each of the interaction term’s constitutive variables, conditional on the level of the other constitutive variable, should also be evaluated (Zelner 2009, 1339).

4. Results of the robustness checks and post hoc analyses are available from the authors upon request.

5. In order to provide relevant and brief support in our discussion section, one author of this paper interviewed one senior manager of Samsung Electronics and one senior manager of LG electronics. The interview questions asked were: (1) ‘Does your company have specific evidence of which Internet infrastructure in host countries leveraged your firm-specific subsidiary advantages by way of connecting your headquarters and subsidiaries ?’; (2) ‘Following the first question, accordingly, does your company have anecdotal evidence that speedy and efficient communications based on ICT strengthened capabilities across borders?’; (3) ‘Does your company have anecdotal cases showing that the development of ICT provides a platform to better access customer preferences and needs?’; and (4) ‘Building on the previous question, does your company have specific evidence regarding which outbound ICT and/or inbound ICT help to improve company’s IT in its supply chains?’