|
|
|
|
Columbia FDI Perspectives
Perspectives on topical foreign direct investment issues
No. 249 April 8, 2019
Editor-in-Chief: Karl P. Sauvant ([log in to unmask])
Managing Editor: Marion A. Creach ([log in to unmask])
|
|
Bilateral investment treaties’ (BITs) capacity to catalyze long-term investment is central to their appeal. However, even after years of study, it is not clear that they encourage investment.
The principal barrier to answering that question is the literature’s practice of proxying long-term investments with balance-of-payments (BOP)-based measures of FDI. Such data capture the net value of capital transactions (debt, equity, reinvested earnings) between MNEs and their affiliates over the course of a year. Their availability for many countries over long periods of time makes them attractive for studying the impact of BITs. However, they are mismatched for the task, and their use is more problematic than commonly understood.
Some of these limitations have been noted in an earlier Perspective.[1] They include:
- Variation in reporting standards, notably regarding the inclusion of reinvested earnings and methods used to value FDI stocks. This limits these data’s usefulness to research designs that require cross-national comparisons, which the study of BITs typically does.
- The association of capital with its immediate source and destination. This can distort information as to where capital is being deployed and by whom when FDI is (as it often is) routed through intermediate destinations and can lead to misidentifying domestic investments as FDI if they are “round-tripped.”
These problems are fixable, and an earlier Perspective describes significant efforts underway.[2] Other features of these data are not as fixable. Most importantly:
- BOP-based FDI data include capital flows associated with efforts to minimize tax liability or take advantage of local interest rates, rather than to serve any long-term enterprise. These flows more closely resemble portfolio investments,[3] and it is unclear whether BITs would affect them. Their inclusion limits these data’s ability reliably to estimate BITs’ effects on the relevant flows.
- BOP-based FDI data exclude the accumulation of assets by MNEs’ foreign affiliates through means other than FDI, including local borrowing. That is not a flaw in the data—local borrowing is not FDI—but when such funds are used to finance risk-sensitive projects, they are relevant to estimating BITs’ potential impacts.
- BITs not only plausibly affect the localization of MNEs’ operations, but also whether these operations are financed in ways that are captured in BOP-based FDI data. Lowered perceptions of political risk can be expected to improve the terms on which affiliates can borrow locally, and encourage them to do so.[4] Because local borrowing is excluded from FDI data, BITs might increase the scale of MNEs’ operations while decreasing its visibility to analysts using FDI data.
These problems are inherent to BOP-based FDI data and cannot really be fixed. A better understanding of BITs’ impact on MNEs’ operations requires measuring these operations more directly.
The most commonly used, publically-available[5] datasets amenable to this purpose are the US Bureau of Economic Analysis’ (BEA) “Activities of U.S. multinational enterprises” data[6] and the OECD’s “Activities of multinational enterprises” data,[7] both of which use firm-level surveys to capture MNEs’ foreign assets and operations, regardless of how those assets and operations are financed. These data are often more directly relevant to the study of BITs. The BEA’s data, for example, allow us fixed capital investment, independent of whether it is financed through local borrowing or through FDI, or if the relevant capital flows were routed through intermediate destinations.
These data are imperfect, of course. The BEA’s data require a singular focusing on the operations of majority-owned US MNEs. The OECD data offer a broader sample, but limited coverage of MNEs’ activities in the developing countries that are most relevant to analyze BITs’ effects. Nonetheless, these data are often preferable to traditional alternatives and should be used whenever research designs allow it. That is a substantial qualification, particularly with respect to sample size, but those limitations are typically preferable to those posed by BOP-based FDI data.
* The Columbia FDI Perspectives are a forum for public debate. The views expressed by the author(s) do not reflect the opinions of CCSI or Columbia University or our partners and supporters. Columbia FDI Perspectives (ISSN 2158-3579) is a peer-reviewed series.
** Andrew Kerner ([log in to unmask]) is a Research Associate in the Political Science department at Michigan State University. This Perspective is based on Andrew Kerner, “What can we really know about BITs and FDI?” ICSID Review-Foreign Investment Law Journal, vol. 33 (2018), pp. 1-13. The author is grateful to Maria Borga, Raymond Mataloni and Theodore Moran for their helpful peer reviews.
[3] Olivier Blanchard and Julien Acalin, “What does measured FDI actually measure?” Policy Brief, 16-17 (Washington: PIIE, 2016).
[4] Selin Sayek, Alexandre Lehmann and Hyoung Goo Kang, “Multinational affiliates and local financial markets,” Working Paper 04/107 (Washington: IMF, 2004).
[5] Commercial datasets are sometimes available as well, albeit at a cost and often without substantial temporal coverage.
|
|
|
The material in this Perspective may be reprinted if accompanied by the following acknowledgment: “Andrew Kerner, ‘How to analyze the impact of bilateral investment treaties on FDI,’ Columbia FDI Perspectives, No. 249, April 8, 2019. Reprinted with permission from the Columbia Center on Sustainable Investment (www.ccsi.columbia.edu).” A copy should kindly be sent to the Columbia Center on Sustainable Investment at [log in to unmask].
|
|
For further information, including information regarding submission to the Perspectives, please contact: Columbia Center on Sustainable Investment, Marion A. Creach, [log in to unmask].
Most recent Columbia FDI Perspectives
- No. 248, Stephan W. Schill and Geraldo Vidigal, “Investment dispute settlement à la carte within a multilateral institution: A path forward for the UNCITRAL process?,” March 25, 2019
- No. 247, Karl P. Sauvant, “The state of the international investment law and policy regime,” March 11, 2019
- No. 246, Joachim Pohl, “Is international investment threatening or under threat?,” February 25, 2019
All previous FDI Perspectives are available at http://ccsi.columbia.edu/publications/columbia-fdi-perspectives/.
Other relevant CCSI news and announcements
- CCSI is hiring a new Special Assistant to the Director to begin in late spring/early summer 2019. To apply for the position, please see our website here for more information.
- CCSI is hiring two Researchers on SDG-aligned Practice in Energy and Agribusiness Sectors. Please see our website here for more information and to apply.
- In February 2019, CCSI released a new Emerging Market Global Players (EMGP) Report: Uncertain Expectations of Mexican-American Economic and Trade Relations and Slowdown of Overall Mexican FDI. The EMGP project, a collaborative effort led by CCSI, brings together researchers on FDI from leading institutions in emerging markets to gather original data from company surveys and additional research and to produce annual reports based on their findings.
|
|
Karl P. Sauvant, Ph.D.
Resident Senior Fellow
Columbia Center on Sustainable Investment
Columbia Law School - Earth Institute
Ph: (212) 854-0689
Fax: (212) 854-7946
|
|
|
|
|
|
|
