Research Subnational EUE & Air Pollution FDI & Climate Adaptability World-System Position & Mobility Generative ABM of Dependency Core Capital & Income Inequality Diffusion of Environmental Discourse Publications Teaching Awards In the News About Contact

Project Overview

1,820Country-Year Observations
152Countries
15Years (2009 to 2023)
+.51CCDI Effect on Gini (M4)

Does foreign direct investment widen or narrow income inequality? The mainstream development literature reports that FDI promotes convergence, but it typically measures FDI as a simple volume or stock. This project argues that what matters is not how much foreign capital a country receives but its structure: where the capital comes from, how core-like its senders are, and how concentrated the dependency is. Using the Core Capital Dependency Index built from IMF bilateral FDI position data, the study estimates the effect of structural dependency on the Gini coefficient across 152 countries from 2009 to 2023.

Income inequality is measured with the SWIID net-income Gini. The analysis uses growth curve models with hierarchical linear modeling, region-specific and OECD curvilinear time trajectories, and cluster-robust standard errors. Across four nested specifications, Core Capital Dependency carries a consistent positive effect on inequality, robust to FDI stock, sector composition, demographic structure, education, trade openness, institutions, and labor-market controls. The effect operates through network structure, not FDI volume.

The Core Capital Dependency Index

The index is a relational measure of FDI that captures the source of foreign capital, the receiver's structural position in the global investment network, and its reliance on foreign capital relative to its own outward investments. It is derived from IMF bilateral FDI position data and recomputed for each country in each year, scaled to a 0 to 1 range.

Construction in Brief

1
Coreness. Countries are ranked each year by total outward FDI position and normalized to a 0 to 1 coreness score, where 1 is the largest outward investor.
Coreᵢₜ = (Nₜ − rankᵢₜ) / (Nₜ − 1)
2
Composition. For each receiver, the share of inward FDI from each sender is weighted by the sender's coreness, scoring higher when inflows come from more core-like economies.
Compᵢₜ = Σ Sᵢⱼₜ × Coreⱼₜ
3
Reliance and the index. Reliance is inflows over total flows; the index rises when a country relies on core capital and is itself less central.
Dependencyᵢₜ = Compᵢₜ × ρᵢₜ × (1 − Coreᵢₜ)

The Global Distribution of Inequality and Dependency

The two maps show the global distribution of income inequality and core capital dependency. The top panel maps the net-income Gini coefficient: the Global North enjoys lower within-country inequality relative to the Global South, with the highest concentrations in southern Africa, where Namibia (64.3), South Africa (61.2), and Botswana (57.5) rank among the most unequal in the world, while Czechia (24.6), Slovenia (23.9), and the Slovak Republic (22.3) show the lowest. The bottom panel maps the Core Capital Dependency Index: high dependency concentrates in sub-Saharan Africa, central Asia, the western Andes, and the post-Soviet region, while western Europe and North America show the lowest scores. Use the buttons to switch panels; hover for country values.

Income Inequality (Gini Coefficient)
Gini (net income) shown as latest available per country through 2023, N = 185; Core Capital Dependency for 2023, N = 168. Dependency from IMF DIPCE bilateral FDI positions; Gini from SWIID. Hover for country values and the Gini reference year.

Estimated Effects on Income Inequality

The coefficient plot presents the full model (M4), displaying point estimates and 95 percent confidence intervals for all predictors. The most prominent feature is the Core Capital Dependency coefficient, which is positive, significant, and among the largest in magnitude relative to the scale of the other predictors. The non-significant FDI stock terms, whose intervals span zero, stand in clear contrast, reinforcing that it is the structural rather than the volumetric dimension of foreign investment that drives the inequality effect. Among the controls, sector pluralism and migrant population are positive and industrial employment negative, while the remaining predictors cluster around zero. Switch between the dependency effect across the four models and the full M4 specification.

Full Model (M4): All Predictors
N = 1,820 country-year observations, 152 countries. Growth curve models with region-specific and OECD curvilinear time trajectories. Cluster-robust SEs. † p < .10; * p < .05; ** p < .01; *** p < .001.

Key Findings

Core Capital Dependency raises the Gini coefficient. The effect is positive and significant in every specification, from 0.483 in the baseline to 0.514 in the fully saturated model, and it survives the inclusion of FDI stock and its square, sector composition, demographic structure, education, trade openness, institutions, and labor-market controls. A one-unit move across the full range of the index is associated with roughly half a Gini point on average, net of everything else.

The result matters because the index measures structure, not volume. Raw FDI stock as a share of GDP carries only a marginal effect, and its square is null, so the inequality consequence of foreign investment is not about how much a country receives. It is about the configuration of who sends it and how concentrated and core-origin that capital is. This directly challenges the convergence claim built on volume measures.

Among the controls, the pattern is coherent with stratification theory: sector pluralism and migrant population share raise inequality, while industrial and agricultural employment lower it. The dependency effect holds alongside all of them, which is the paper's central empirical claim: structural dependency on core capital is an independent driver of income inequality.