James Thurlow

An agentic AI assistant for country-level economic modeling: Methods, data, and expert evaluation

The demand for high-quality, rapid economic analysis to navigate complex issues faced by many low- and middle-income countries has led to the development of detailed structural simulation models, such as Computable General Equilibrium (CGE) models. Policy analysis with such models requires deep knowledge of their structure and applicability to the policy issues at hand. Policymakers in these settings often lack access to the expertise required for articulating, analyzing, and interpreting the relevant causal chains captured by the models. Attempting to circumvent these barriers by submitting complex economic questions directly to off-the-shelf large language models (LLMs) introduces severe analytical risks, including hallucinations and insufficient expert guidance. To resolve this limitation, we developed and empirically evaluated an agentic AI assistant called RIAPA-AI that integrates LLMs with a CGE model. We evaluated the performance of RIAPA-AI against expert human CGE modelers and a general-purpose plain LLM baseline across samples of complex economic scenarios, utilizing an independent panel of senior economists to grade the outputs. Our statistical analysis reveals no statistically significant difference in analytical accuracy between RIAPA-AI and human experts, while the AI accelerates reproducible policy analysis from weeks to minutes. Furthermore, by operating without manual processing limits, RIAPA-AI eliminates the 6.7% error rate observed among human modelers. Conversely, the general-purpose plain LLM exhibits profound failure rates, failing to achieve policy-ready scores in over 60% of depth evaluations. Without an underlying CGE model acting as a bounding force to reflect economic structural constraints, the general-purpose plain LLM defaults to linear economic assumptions and inserts unmodeled socio-political narratives. Crucially, by explicitly restricting the AI’s narrative interpretation solely to deterministic numerical outputs, RIAPA-AI mitigates the risk of unverified assumptions and logic hallucinations. We conclude that by deploying an agentic AI assistant that layers a generative AI over a formal CGE model, RIAPA-AI successfully delivers sensible, rigorous, and rapid policy analysis.

Agrifood systems of India: A state-level analysis using a social accounting matrix

Gender, youth, and growth: Unpacking the productivity–transformation nexus in Rwanda

The relationship between labor productivity and economic transformation, and their combined impact on labor market dynamics, remains insufficiently understood and highly country specific. This study applies an economywide analytical framework, using Rwanda as a case study, to examine how youth and women’s productivity influence economic growth and structural transformation, and how this transformation process, in turn, affects these groups. The results indicate that labor productivity gains—whether economywide or concentrated among youth—shift growth toward the industrial and service sectors, while growth in the agricultural sector is minimal. Increases in productivity of women’ labor generate more balanced growth across sectors and substantially enhance women’s industrial participation. Productivity gains of youth labor induce stronger structural shifts, as young workers move from agriculture to expanding industrial and service sectors, though this transition partially displaces adult workers. In general, labor income in these simulations rises broadly in line with GDP, with youth and women benefiting most under targeted scenarios. Sector-specific growth strategies yield distinct distributional effects, however: industry-led growth benefits women and adults, while service-led growth favors the versatile youth. Overall, productivity-driven structural transformation in Rwanda fosters welfare gains, although potential trade-offs between inclusiveness across gender and age groups and aggregate economic performance warrant further investigation. In conclusion, policy design in Rwanda should ensure that gains in aggregate economic growth are balanced with inclusive outcomes for women, youth, and adults.