Foreign investment lifts sustainability unevenly across Vietnam’s provinces

Foreign direct investment has long been promoted as a universal engine of development, but a new study of Vietnam’s provinces suggests that its benefits are anything but uniform. Analyzing a balanced panel of 63 Vietnamese provinces from 2019 to 2022, researchers report a positive association between implemented FDI and a composite measure of provincial sustainable development. Yet the strength of that association shifts markedly with local conditions: it grows stronger where domestic development investment is higher and weaker where multidimensional poverty is more prevalent.

The study, published in Environmental and Sustainability Indicators, is notable for moving beyond the country-level analyses that dominate the FDI literature. Instead of asking whether foreign investment helps or harms nations in aggregate, the authors examined how the relationship plays out within a single country, where provinces differ enormously in infrastructure, income, poverty, and economic structure. They used the Provincial Sustainable Development Index, or PSDI, a composite score developed since 2019 by Vietnam’s Ministry of Planning and Investment with technical support from the United Nations Development Programme. The index applies an equal-weighting approach and, because of data availability, covers 14 Sustainable Development Goal components rather than the full set of 17.

The dataset yielded 252 province-year observations. Implemented FDI inflows, measured in millions of US dollars and drawn from provincial statistical yearbooks and the General Statistics Office of Vietnam, ranged from zero to roughly 22.4 billion dollars, with a mean of about 542.63 million and a standard deviation of 2,604.74 million. The PSDI ranged from 28.51 to 66.46 points, averaging 50.92. The authors chose implemented rather than registered FDI because realized capital better reflects actual investment activity; over 2010 to 2022, implemented FDI averaged roughly 60 percent of registered commitments, and the conversion rate varied considerably across provinces.

The core analysis used a two-way fixed-effects model with province and year fixed effects and standard errors clustered at the province level. In the preferred specification, the coefficient on log-transformed FDI was 0.106, statistically significant at the one percent level, indicating that provinces with higher FDI inflows tend to show higher sustainable development scores after accounting for controls including gross regional domestic product, forest coverage, treated solid waste, and environmental tax revenue. The authors are careful to frame this as a conditional within-province association rather than a definitive causal effect, given the short four-year panel.

Disaggregating the index revealed a sharper picture. The positive association was concentrated in the economic dimension, with a coefficient of 0.197 significant at the one percent level, and the social dimension, with a coefficient of 0.135 significant at the five percent level. For the environmental dimension, the coefficient was negative and statistically insignificant, offering no evidence of a measurable relationship. The authors note that FDI-related activity may be more directly reflected in economic and social outcomes, while such gains do not necessarily translate into environmental improvements, though their estimates do not identify the underlying mechanisms.

The moderating results carry the study’s central message. When FDI was interacted with development investment, the interaction coefficient was positive and significant at the ten percent level, supporting the idea that domestic capital formation in infrastructure and public facilities complements foreign investment. When FDI was interacted with the multidimensional poverty rate, which captures deprivations in income, health, education, living standards, and access to basic services, the interaction coefficient was negative and significant at the one percent level. In plainer terms, the same foreign investment appears to travel further in provinces that have built local capacity, and less far where socioeconomic constraints are deeper.

Subgroup analyses reinforced this unevenness. Classifying provinces by average per capita income, the positive association was statistically detectable among low-income provinces but not clearly so among high-income provinces under either an 80th or 75th percentile threshold. One interpretation offered by the authors is that low-income provinces have greater scope to absorb additional capital, technology, and market access, while wealthier provinces may already possess developed infrastructure that makes marginal gains less pronounced. They caution that these are possible explanations rather than directly tested mechanisms, and they suggest the pattern may even mean FDI reinforces existing disparities rather than promoting convergence.

Regional results added further texture. Following the classification in Vietnam’s 2017 Law on Planning, the 63 provinces were grouped into six economic regions. The FDI coefficient was negative and significant at the ten percent level in the Red River Delta, positive and significant in the North Central and South Central Coast and in the Central Highlands, and statistically insignificant in the Northern Midlands and Mountainous Areas, the Southeast, and the Mekong River Delta. The negative Red River Delta result may reflect the region’s high urbanization and industrial concentration, where congestion and environmental pressures could limit gains in the composite index, though the authors again treat this as interpretation rather than established fact.

Robustness checks strengthened the baseline finding. Replacing implemented FDI with the total number of FDI projects, a measure of the extensive margin of investment, produced positive and significant coefficients in both specifications. The authors also employed an instrumental variable approach using a shift-share, or Bartik, instrument that combines provinces’ pre-2017 industrial exposure to FDI-intensive industries with subsequent global greenfield FDI growth excluding Vietnam. The instrument was strongly predictive in the first stage, identification diagnostics exceeded conventional weak-instrument reference thresholds, and the second-stage estimate of 0.642 was positive and significant for FDI inflows, though results were less conclusive for project counts. Because the exclusion restriction cannot be tested directly, the authors present the IV estimates as complementary evidence rather than definitive causal proof.

The study’s period carries an important caveat: 2020 through 2022 overlapped with the COVID-19 pandemic, which disrupted both FDI flows and development conditions, so the estimates may not represent relationships in normal times. Newly registered FDI projects peaked at 6,842 in 2019, fell during the pandemic, and recovered to 3,571 in 2022. The authors also acknowledge that the aggregate index may conceal differences among its underlying dimensions, that the short panel limits causal inference, and that firm- or sector-level data would be needed to trace channels such as employment, technology adoption, and spillovers to domestic firms. For policymakers, the implication is direct: attracting foreign investment alone is not a development strategy. The findings suggest that FDI attraction policies work best when paired with local development investment and poverty reduction, and that a single national FDI playbook is unlikely to deliver uniform sustainable development outcomes across such a diverse landscape of provinces.

Subject of Research: The association between foreign direct investment and provincial sustainable development in Vietnam, including moderating roles of development investment and multidimensional poverty.

Article Title: Uneven gains from foreign direct investment: Evidence on sustainable development across Vietnamese provinces

Article References: Dong, V.-C., Nguyen, T. D., Pham, P. M. H., & Hoang, V. H. (2026). Uneven gains from foreign direct investment: Evidence on sustainable development across Vietnamese provinces. Environmental and Sustainability Indicators, 32, Article 101584. https://doi.org/10.1016/j.indic.2026.101584

Image Credits: AI Generated

DOI: 10.1016/j.indic.2026.101584

Keywords: foreign direct investment, sustainable development, Vietnam, provincial analysis, multidimensional poverty, development investment, Sustainable Development Goals, panel data, instrumental variable, regional inequality, economic regions, Provincial Sustainable Development Index