The determinants of firms’ global diversification decisions

ABSTRACT

Why do some firms choose to be ‘born global’ and become early internationalizing firms (EIFs) while others choose to be ‘born-again global’ and develop into late internationalizing firms (LIFs)? What are the main factors impacting a firm’s decision on the timing of global diversification? Based on the theories of diversification, this study examines the role of peer influence and desire for growth on the timing of a firm’s globalization decision. We further study the idiosyncratic risk and the adoption of technological innovation hypotheses on global diversification. Our results document that innovation efficiency strongly enhances EIFs’ propensity to global diversification. On the other hand, peer pressure and idiosyncratic risk level significantly influence EIFs not to globalize. In contrast, LIFs are positively influenced by their industry peers, showing how young and mature companies respond to the market competition in a different manner.

KEYWORDS:

• Global diversification
• determinants
• born-global
• born-again global

JEL CLASSIFICATION:

• F23
• F30
• G31
• G32

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ In general, according to Marinelli (2011), firms start their global diversification path by importing and exporting, progressively increasing their internationalization, while some authors argue that the development of exports does not suffice for a firm’s inclusion among global diversified enterprises.

² The root source of the born–again terminology comes from the management and entrepreneurship literature that recognizes the existence of this second internationalization pathway (Bell, McNaughton, and Young 2001; Bell et al. 2003).

³ According to Bernardo and Chowdhry (2002), these specific resources must be easily accessible if substantial hurdles are to be prevented.

⁴ Krishnaswami and Subramaniam (1999) find that organizations with higher growth opportunities are more likely to take part in spin-offs.

⁵ Seru (2014) measures innovation with patent-based metrics and finds that firms acquired in diversifying mergers have lower levels of innovation and even less creative innovation.

⁶ Reduction of the CEO’s human capital vested in the firm is another central idea (Amihud and Lev 1981; May 1995): managers pursue diversification as a result of their personal risk reduction preferences. The ‘managerialism’ findings by Amihud and Lev (1981) emphasize that the risks associated with firms’ performance are also hazards for CEOs. They show that since CEOs’ compensation is tied to firms’ accomplishments, CEOs display a higher interest in conglomerate mergers. May (1995) proposes that managers with more human capital vested in the firm display a higher affinity for risk reduction. May (1995) proxies the CEO human capital vested in the firm by tenure (number of years spent at that firm) and reports that CEOs who have been employed for many years have a greater tendency to diversify.

⁷ We collect financial statements data for IPO firms from 1992 through the end of 2017. However, our sample period ends in 2014 because we need four consecutive years of pre-tax foreign income in order to qualify a firm in 2014 as a global firm. That is, a 2014 global firm needs to have pre-tax foreign incomes in 2014, 2015, 2016 and 2017.

⁸ Research and development spending per total assets proxy for production and marketing in Morck and Yeung (1998) and geographic diversification adds to shareholders value only in the presence of these R&D related assets.

⁹ Knight and Cavusgil (2004) argue that three years or less should be the time frame from domestic creation to initial foreign market entry. We go beyond and generate a sample labelled EIFs (47) which include Div = 1 from year 4 through 7 after the IPO year. The sample does not have enough observations and thus is dropped from further analysis.

¹⁰ To preserve the space the correlation matrix of LIFs is not reported but is available from the authors upon request.

¹¹ Santalo and Becerra (2008) demonstrate that diversified firms perform better in certain industries.

¹² Peer influence operates in the same way for samples EIFs 14 and EIFs 36, when we include year-fixed effects in the regressions and when we are controlling for characteristics specific to a certain industry.

¹³ To preserve the space, results are not reported here but are available from the authors upon request.