Evolution of FDI flows in the global network: 2003–2012

ABSTRACT

As one type of international capital flow, FDI maintains its important role in globalization. This article attempts to investigate the evolution of the FDI flows from a network perspective. Based on the bilateral FDI flows data between countries from 2003 to 2012, we construct the global FDI flows network for each year and thus quantify network measures (such as flow volumes and connections); further by analysing the tendency and changes of these network measures during the past 10 years, we delineate the features and dynamics of the FDI flows in the global network. We have the following findings: (a) the flows network changes during and after the crisis, i.e. flow volume fallen down and recovered, and flow connection restructured with more diversity; (b) the global FDI flows network is getting more loosely connected; (c) individual countries vary in different patterns.

KEYWORDS:

• FDI
• bilateral flows
• network analysis

JEL CLASSIFICATION:

• C82
• D85
• F21

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ http://unctad.org/en/Pages/DIAE/FDI%20Statistics/FDI-Statistics-Bilateral.aspx. The website provides bilateral FDI data for 206 countries from 2001–2012, categorizing into inflows (inward FDI flows), outflows (outward FDI flows), instock (inward FDI stock), and outstock (outward FDI stock). Flows are the annual changes and stocks are accumulating positions.

² Since the country j is the reporting country, outflow is the outward FDI from this country to the other countries.

³ On the occasion when reporting countries use the same standard and statistics without mistakes or omissions, the net bilateral inflow and outflow between country i and country j should ideally equal in absolute value. Besides, for the reporting country, the positive outflows are recorded as negative ones in the data matrix to distinguish them from inflows and thus the latter two terms in the equation can be considered as the opposite value of the outflows from country i to country j.

⁴ This decade covers three periods: The pre-crisis period is from 2003 to 2006; the mid-crisis period is from 2007 to 2009; then the post-crisis period is form 2010 to 2012.

⁵ The top 500 net flows are also filtered by this step and the remained number varies from 423 ~ 445.

⁶ Also refer to Newman (2003) for formal definition about APL and clustering coefficient.

⁷ Centrality measures such as closeness and betweenness are also analysed (not shown for brevity) and they show the decreased value in closeness and the increased value in betweenness, consistently indicating a loosely connected network.

⁸ By comparing net inflows (outflows) in 2008 and 2009, we find that inflows decreased in 44 countries while outflows decreased only in 28 countries; meanwhile, inflows increased in 22 countries while outflows increased in 36 countries. The aggregate net flows of 68 countries fell down $244,949 million.

⁹ Similar with flow volume, there were 35 countries that received FDI from less partner countries while there were only 21 countries from more countries in 2009. Meanwhile, 34 countries increased outward FDI partner countries and only 19 countries decreased their outward partners.

Additional information

Funding

This work was supported by the Natural Science Foundation of Guangdong Province, China [2016A030313744].