Working Papers

Absorptive Capacity and Total Factor Productivity Effects of FDI

Abstract: This paper investigates how domestic absorptive capacity conditions the impact of foreign direct investment (FDI) on total factor productivity (TFP) growth in developing economies. Constructing a multidimensional index of human capital, R&D, and financial development that enforces essential complementarity among structural capabilities, I estimate a simultaneous-equations system for 59 developing countries over the 1996–2018 period. The results reveal a clear productivity gradient: FDI and absorptive capacity act as structural complements. While FDI exerts a negative direct effect on TFP growth, this penalty is substantially mitigated by a positive and highly significant interaction with capacity, which acts as a productivity multiplier. Moving from the tenth to the ninetieth percentile of capacity reduces the productivity drag associated with a one-percentage-point rise in the FDI-to-GDP ratio from 0.28 to 0.01 percentage points.

Although this attenuation effectively neutralizes the penalty in high-capacity environments, I find no evidence of a statistically significant productivity premium within the observed sample. This pattern is unique to productivity; when real GDP growth is the outcome, the interaction with capacity reverses sign, and returns to FDI are instead conditioned by distance to the technology frontier. These findings suggest that inward FDI is not a universal catalyst for productivity; absent sufficient domestic foundations, the mismatch between local capabilities and frontier technology prevents FDI from delivering expected productivity gains.

Resource Nationalism and Domestic Value Capture: Effects of Export Bans on FDI and Processing-Sector Employment in Resource-Rich Economies,” with Craig Parsons.

Abstract: A coercive resource nationalism has emerged across critical mineral supply chains: governments ban raw ore exports to compel domestic downstream processing. We evaluate whether this “forced downstreaming” generates domestic value capture or merely shifts customs statistics while creating capital-intensive foreign enclaves. Combining the OECD Inventory of Export Restrictions for 60 resource-rich economies (2003–2024) with product-level trade, industrial employment, and greenfield FDI data, we apply heterogeneity-robust difference-in-differences, synthetic difference-in-differences, and randomization inference. We show that product scope governs policy efficacy. Only 2% of recorded mineral export bans target raw ores; most restrict scrap, slag, or base metals.

These broad bans have no detectable effect on processed export shares, basic metals employment, or processing FDI. By contrast, bans on unprocessed nickel, cobalt, bauxite, and lithium ore—enacted by Indonesia, DR Congo, and Zimbabwe—raise the processed export share by 41 percentage points (randomization p = 0.008). A within-mineral decomposition reveals that this shift is driven primarily by the collapse of raw ore shipments rather than a significant expansion in processed exports. In Indonesia, the sole ore-ban adopter with disaggregated employment data, formal smelting employment rose 2.5-fold relative to never-banning producers alongside US$34 billion in announced Chinese processing investment, while battery manufacturing employment remained stagnant. Raw ore bans can relocate the initial processing tier in producers with market power, but they have not, so far, pulled later stages of the value chain with them.

Does Trade Fragmentation Insulate Emerging Markets from U.S. Monetary Policy?

Abstract: Since 2018, tariff escalation and geopolitical realignment have fundamentally redirected global trade. This paper examines whether this structural reorganization has altered the cross-border transmission of U.S. monetary policy to emerging economies. Using a quarterly panel of 43 emerging economies (2000–2024) and high-frequency Federal Reserve surprises purged of predictable news, I estimate state-dependent local projections evaluating transmission through U.S. import demand, currency invoicing, commodity dependence, and dollar liabilities. I isolate trade diversion using a shift-share design based on the product-level Section 301 tariffs imposed on Chinese goods in 2018–2019, alongside trade data separating productive reallocation from third-country rerouting.

The trade channel is empirically weak: U.S. export exposure does not amplify output spillovers, and neither dollar invoicing nor commodity dependence generates robust differential macroeconomic responses. By contrast, the financial channel dominates: dollar liabilities significantly amplify long-term sovereign bond yield responses, an effect concentrated in 2020–2024. Crucially, post-2018 trade diversion has not decoupled emerging markets: economies most exposed to diversion saw ten-year yield sensitivity increase by 2.0 percentage points after eight quarters for a 25-basis-point tightening, with no robust change in output.

This heightened sensitivity is not driven by the rerouting of Chinese goods and survives excluding 2016–2017. The increase tracks exposure to the tariffed products: exposure to goods slated for tariffs that never took effect predicts no change, though the data cannot tie the effect to individual tariff lines. Trade fragmentation has not insulated emerging markets; rather, China substitutes have become more financially exposed, underscoring the primacy of balance-sheet over trade-based buffers.

Tariffs and the Dollar: Does Trade Fragmentation Lead to Financial Fragmentation?

Abstract: For decades, the global primacy of the U.S. dollar has been anchored by its central role in international trade. This paper investigates whether recent geoeconomic fragmentation in trade engenders a parallel fragmentation in global finance. To operationalize this inquiry, I construct a novel country-level Dollar Financial Fragmentation Index (DFFI) based on the currency composition of cross-border bank claims and international bonds, covering 151 economies quarterly through 2026Q1. The index reveals that while financial fragmentation away from the dollar is measurable, it is remarkably narrow: since 2017, the non-dollar share of cross-border bank claims has risen by six percentage points at constant exchange rates, whereas bond finance has consolidated toward the dollar.

The largest country-level shifts are driven by sanctioned economies, euro-linked jurisdictions, and China, rather than by geopolitical distance from the United States. A broader annual Financial Fragmentation Index (FFI)—incorporating gold reserves, trade invoicing, geopolitical portfolio bias, and Chinese swap-line access—demonstrates that fragmentation margins are nearly uncorrelated within countries. Evaluating these indices against the 2025–2026 universal U.S. tariff threats and the 2018–2019 U.S.–China trade war yields precise null results for the DFFI: neither the threat of trade fragmentation nor its realized reallocation detectably moved countries away from the dollar.

Component analyses using TIC and BIS data confirm these aggregate findings. Following the 2018 shock, exposed emerging economies saw a pullback by U.S. portfolio investors, larger in equity and consistent with benchmark rebalancing, but exhibited no durable shifts in dollar-denominated debt, currency invoicing, or U.S. Treasury demand. Ultimately, the data reject the financial coercion hypothesis, underscoring the inertia of the dollar’s financial network effects.