Sample Answer
Overview and main contribution
This paper examines whether foreign direct investment (FDI) generates knowledge diffusion in economically poor locations, and if so, through which mechanisms such diffusion occurs. The central question is both theoretically and policy relevant, as many developing countries actively pursue FDI through incentives under the assumption that foreign firms transfer knowledge, skills, and productivity gains to local economies.
The authors combine firm-level data with spatial and sectoral information to analyse productivity spillovers from foreign-owned firms to domestic firms operating in close proximity. The paper distinguishes between different channels of diffusion, including worker mobility, supplier and buyer linkages, and demonstration effects. The focus on poor locations is a key contribution, as much of the existing literature concentrates on middle-income or rapidly industrialising economies rather than structurally disadvantaged regions.
Overall, the paper aims to challenge the assumption that FDI automatically leads to local learning and development, arguing instead that the extent of knowledge diffusion depends on local absorptive capacity, institutional quality, and the nature of foreign investment.
Importance and relevance
The topic is important for several reasons. First, it contributes to the long-standing debate on whether FDI should be viewed as a development tool or merely as a source of enclave growth. Second, it speaks directly to policy design in low-income regions where governments often sacrifice tax revenue and regulatory autonomy to attract multinational firms. Third, by focusing on knowledge diffusion rather than aggregate growth outcomes, the paper addresses a more nuanced and mechanism-based understanding of development impacts.
The emphasis on poor locations is particularly valuable. The literature has increasingly acknowledged that spillovers are not automatic, yet empirical evidence from low-capacity environments remains limited. By attempting to fill this gap, the paper has the potential to influence both academic debates and applied policy discussions.
That said, the paper’s contribution would benefit from a clearer positioning relative to existing spillover studies. While several relevant strands of the literature are cited, the manuscript could do more to explicitly state what is genuinely new beyond applying established methods to poorer contexts.
Conceptual framework and hypotheses
The conceptual framework is grounded in standard theories of knowledge spillovers and absorptive capacity. The authors correctly note that foreign firms may possess superior technologies, management practices, and organisational routines, but that diffusion requires interaction and learning on the part of domestic firms.
The discussion of absorptive capacity is appropriate but somewhat underdeveloped. While the paper acknowledges the importance of human capital, firm size, and institutional quality, these factors are not always clearly integrated into the empirical hypotheses. In several sections, absorptive capacity appears more as an ex post explanation than as a core element guiding empirical design.
The hypotheses would benefit from sharper formulation. For example, the paper could more explicitly distinguish between conditions under which spillovers are expected to occur versus conditions under which FDI may crowd out domestic firms or operate in isolation. This would strengthen the analytical coherence of the study and help readers better interpret the results.
Data and methodology
The use of firm-level data is a major strength. The dataset appears rich and allows for disaggregation across sectors, ownership types, and geographic proximity. The authors make a credible effort to identify spillovers by exploiting spatial variation in exposure to foreign firms.
However, several methodological concerns merit closer attention.
First, endogeneity remains a significant challenge. Foreign firms do not locate randomly, especially in poor regions. They may choose locations with better infrastructure, more capable firms, or favourable local institutions. While the paper includes controls and fixed effects, it is not always clear that these fully address selection concerns. The identification strategy would be stronger if the authors more clearly justified why remaining bias is unlikely to drive the results.
Second, the measurement of “knowledge diffusion” relies heavily on productivity outcomes. While this is common in the literature, it is also indirect. Productivity changes can result from multiple factors unrelated to learning, including competitive pressure or market reallocation. The paper would benefit from a more explicit discussion of this limitation and, if possible, supplementary evidence using alternative indicators such as innovation activities, training, or worker mobility.
Third, the treatment of distance and proximity deserves further clarification. While geographic closeness is a reasonable proxy for interaction, the paper could more clearly explain why the chosen spatial thresholds are appropriate, especially in poor locations where infrastructure constraints may limit effective interaction even at short distances.
