Foreign Direct Investment and Host-Country Capital Formation: Measurement–Estimand Alignment, Entry Modes, and the Robustness of Inference in a Global Panel
Year: 2026
Type: preprint
Abstract: Foreign direct investment (FDI) is often treated as a homogeneous source of external finance, although announced greenfield projects and cross-border mergers and acquisitions (M&A) represent economically distinct transactions, and prior evidence on their relationship with host-country investment remains contradictory. This study examines whether these entry modes have statistically discernible conditional associations with host-country gross fixed capital formation and whether these inferences remain stable across empirical designs that preserve or alter the underlying estimand. The empirical warehouse combines UN Trade and Development World Investment Report 2026 statistical annexes with World Development Indicators for 2003–2025. The preferred dynamic specification uses 3,111 country-year observations from 162 economies over usable estimation years 2004–2024, with country fixed effects, region-by-year shocks, macroeconomic controls, and lagged capital formation. A same-estimand audit varies common-shock treatment, estimator, weighting, and internal-instrument design, while broader checks alter timing, functional form, common-factor treatment, and outcome definition. Capital formation is strongly persistent (ρ = 0.700), but raw GDP-scaled greenfield and M&A coefficients are statistically imprecise in the preferred model and remain so across diagnostically retained same-estimand GMM designs. A positive greenfield coefficient appears under an inverse-hyperbolic-sine transformation and in one alternative System-GMM specification, but the former changes the functional-form estimand and the latter rejects a relevant Difference-in-Hansen subset test. The findings therefore support measurement–estimand discipline and harmonized robustness inference rather than a universal entry-mode ranking, implying that policy evaluation should distinguish investment form, measurement, and realized productive-capacity outcomes rather than infer development effects from an aggregate FDI label alone.
Source: Research Square
Author: Nitiphong Songsrirote
Institution: Mahasarakham University
Language: English
Subfield: Strategy and Management
Domain: Social Sciences
Open Access status: green
Funder: Mahasarakham University
OpenAlex ID: W7236410445
