Correction

This article refers to:

Outward Foreign Direct Investments as a Catalyst of Urban-Regional Income Development? Evidence from the United States

Article title: Outward Foreign Direct Investments as a Catalyst of Urban-Regional Income Development? Evidence from the United States

Authors: Harald Bathelt and Maximilian Buchholz

Journal: Economic Geography

Citation details: Volume 95, Number 5, pages 442 - 466

DOI: https://doi.org/10.1080/00130095.2019.1665465

The authors regret to report that, due to an error, Tables 1, 2, and columns 4 and 5 of Table 3—in which they explore the determinants of median income in US Combined Statistical Areas (CSAs)—do not include time fixed-effects but only CSA fixed-effects (see original and corrected results in attached table). The authors believe that this does not question the findings in these tables, especially since they constructed their investment variables (accumulated over three years normed by population size) in a way to reduce potential for unrelated linear trends in both the dependent and core independent variables. Also, extensive robustness tests with various pooled regression models, two-stage least square (2SLS) and generalized method of moments (GMM) instrumental variables estimators are unaffected and strongly support the findings, particularly the significant positive effect of outward greenfield foreign-direct investments (OGFDIs) on median income in US CSAs.

In checking panel regression models with both time and CSA fixed-effects, the authors were able to support their findings but found some interesting deviations that highlight the need for more future research. In Table 2, the results for OGFDIper10,000 remain once time fixed-effects are included. Regarding Table 1, the analysis for the time period after the global financial crisis from 2010 to 2013 provided strong support for the original findings, particularly the positive and significant impact of OGFDIper10,000, HighWageIndustryEmpShare, and CollegeDegreeShare on ln (MedianIncome). The authors reservations regarding the role of IGFDIper10,000 were also confirmed. For the time period before the financial crisis from 2005 until 2008, the results were different however. Importantly, OGFDIper10,000 was insignificant.

The authors suspect that this different relationship between outward investments and median income in CSAs may be related to the financial crisis and nature of investments. While many investments in the early period may have had the character of relocations of facilities and offshoring, recent investments may be more in line with the opportunity-searching and knowledge-based investments emphasized in their conceptualization. What this suggests is that we are only at the beginning of exploring the relationship between economic networks of cities and their prosperity levels and that we need to understand much better how these relationships develop and change over time.

Key Original and Corrected Results

	CSA Fixed Effects All Years (Original Results)		CSA and Year Fixed Effects 2005–08		CSA and Year Fixed Effects 2010–13		CSA and Year Fixed Effects All Years
	Base Model	Full Model	Base Model	Full Model	Base Model	Full Model	Graduate Degree	Gini	Unemployment
CollegeDegreeShare	0.015***	0.013***	0.008***	0.008***	0.006***	0.006***	0.004	0.001	-0.096
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.001)	(0.068)
HighGrowthIndustryEmpShare	0.100***	0.087***	0.068	0.064	-0.057	-0.055	-0.185**	-0.037***	-2.409
	(0.023)	(0.023)	(0.062)	(0.062)	(0.041)	(0.040)	(0.074)	(0.013)	(1.539)
HighWageIndustryEmpShare	0.360*	0.376*	0.236	0.234	0.170**	0.209***	0.091	-0.095***	-7.746*
	(0.194)	(0.195)	(0.211)	(0.211)	(0.076)	(0.079)	(0.131)	(0.030)	(4.488)
ln(Population)	0.168***	0.154***	0.014	0.006	-0.014	-0.004	-0.042	-0.013	1.404
	(0.036)	(0.033)	(0.055)	(0.055)	(0.022)	(0.023)	(0.029)	(0.011)	(1.157)
OGFDIperCapita		0.054***		-0.058		0.043***	0.043**	-0.005	-0.888
		(0.019)		(0.039)		(0.014)	(0.021)	(0.008)	(0.980)
IGFDIperCapita		0.126***		0.014		0.064	-0.02	-0.007	-0.061
		(0.040)		(0.070)		(0.046)	(0.059)	(0.011)	(1.455)
Adj. R2	0.261	0.279	-0.261	-0.260	-0.297	-0.283	-0.118	-0.109	-0.116
Observations	860	860	382	382	382	382	860	767	860

Notes: Fixed-effects regressions on dependent variable ln(MedianIncome) except in the last three columns where the dependent variables are ln(MedianIncome) for graduate degree holders in a CSA, a CSA’s Gini coefficient and a CSA’s unemployment rate, respectively. Significance level: *** p < 0.01; ** p < 0.05; * p < 0.1. Heteroscedasticity and autocorrelation consistent standard errors are reported in parentheses. Data sources are listed in Table A.1 in Online Appendix A.