![Sample our Economics, Finance,Business & Industry journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5931/TOC-Subject-Sample-2021-Economics-Finance-Business-Industry.jpg"})
Abstract
While the effect of foreign direct investment (FDI) on economic growth at the aggregate level is ambiguous, recent studies at the sector level suggest that this could be due to aggregation. And yet, studies on the effect of FDI on growth at the sector level are scant in the literature. In this paper, I empirically examine the effect of FDI on growth using detailed data on worldwide mergers and acquisitions (M&A) activity at industry level from 1986 to 2010. My results show that foreign acquisition has positive effect on the host country’s economic growth overall. However, the effect varies by sector and foreign acquisition into extractive industries has least positive effect on growth. This also seems to be more pronounced in natural resource-abundant developing countries. On the other hand, foreign acquisition into manufacturing sector has biggest positive effect on growth.
Disclosure Statement
No potential conflict of interest was reported by the author(s).
Notes
1 For example, oil and gas extraction and mining industries.
2 Other forms of FDI include greenfield FDI (establishment of new productive asset with sole ownership) and joint venture (establishment of new productive asset with joint ownership).
3 This becomes more pronounced in developing countries where natural resources are abundant.
4 I use 10% as the threshold level because an investment by an MNC is considered FDI if MNC establishes or purchases an ‘effective voice in management’ of productive assets in another country. And ‘effective voice in management’ is defined as having 10% or more ownership. The regression results are qualitatively identical when 20% or 50% is used instead as the threshold level. These results are available upon request.
5 I use World Bank’s classification of low-income, low-middle income, upper-middle income, and high-income countries for year 2000.
6 For few instances where it was not possible to do five-year average due to missing data, I did four-year, three-year, or two-year average instead.
7 Also, value-added data at the industry-level, especially at the disaggregated level are not available for many countries in my sample.
8 Link to this data source: https://data.worldbank.org
9 All the control variables are in five-year intervals as well.
10 I also tried logged initial real GDP per capita instead as a robustness check and got qualitatively very similar results.
11 My IV is in five-year intervals as well, cumulated into five five-year periods.
12 I could not perform over-identification test because I do not have more instruments than endogenous regressors.
13 The five-year average annual growth rate is multiplied by 100.
14 Each column is not a separate regression. All the interaction terms are included in one single equation and estimated simultaneously. The coefficient estimate of the interaction terms are shown in different columns for each sector for the purpose of easier comparison.
15 The coefficient estimate for the triple interaction term for the manufacturing sector is not significant. But the combined coefficients of the interaction terms are jointly significant.
16 But the combined coefficients of the interaction terms are jointly significant. The coefficients on the interaction term are significant for all the other three industries.
17 I used the year 1988 for Aruba and the year 1989 for Brunei due to missing data in the year 1987 for these two countries.
Additional information
Notes on contributors
Donghyun Lee
Donghyun Lee is an Assistant Professor of Economics who writes on foreign direct investment and cross-border mergers and acquisitions in the presence of firm heterogeneity and multinationals.