Working Papers
Vocational Interests and Work Context in the Choice of Occupational Destination
Abstract:
I study what predicts a worker’s realized occupational destination when capability measures make many moves look feasible. I combine linked Current Population Survey occupation changes with Occupational Information Network occupation attributes and estimate a survey-weighted conditional logit in which each mover chooses among the realized destination and sampled alternatives. Capability defines a large feasible set: about 76 percent of occupation pairs in the top decile of work-activity similarity never exchange a worker, and a broader content screen that admits essentially all realized moves contains roughly ten times as many feasible pairs as realized flows. To a capability, co-location, and wage baseline, I add two occupation-pair measures: the congruence of origin and destination vocational-interest profiles and the similarity of their work-context profiles. Both are measured at the occupation level; individual interests and working-condition preferences are not observed. Adding these channels raises held-out top-one accuracy from 0.279 to 0.338 and McFadden pseudo-R² from 0.188 to 0.243. The lift survives occupation-out validation, choice-set redraws, a mixed-logit relaxation of independence of irrelevant alternatives, and an adjacent-month transition construction. It carries a clear scope condition: among near-substitute occupations that O*NET already flags as skill-related, the top-one lift compresses to about 1.4 points. The estimates are descriptive. They point to non-capability inputs that a defensible estimate of mobilizable labor supply should consider, especially where candidate occupations are reachable but not close substitutes.
A working draft is available at: https://ssrn.com/abstract=7066198
Trade in the Spotlight: Enhancing Gravity Model Predictions with Nightlights and Population-Weighted Distance Measures
Co-Author: Ian Helfrich
Abstract:
The distribution of consumers and producers across and within countries significantly influences the pattern of economic production. In this paper, we investigate the implications of spatial factors on trade model estimates by introducing a novel geodesic distance measure between countries, using centroids weighted by nightlights and by population distributions as endpoints for bilateral distance measures. Traditional gravity models of international trade have relied on time-invariant measures of bilateral distance between nations, including distance between national capitals, geographic centroids, and weighted centroids based on the largest population centers. By employing annual global population and nighttime lights density rasters to identify the central spatial tendency of producers and consumers, we estimate time-varying directional distance measures between origin-destination pairs. Our approach demonstrates increased accuracy and adaptability across various administrative boundaries, offering improved comparability across different research applications. We present the application of this method to interstate commerce in the United States and global trade flows, highlighting the flexibility and reliability of our approach for various empirical considerations.
A working draft is available at: https://ssrn.com/abstract=5202676
Build It and They Will Come? US Regional Labor Composition and Readiness to Meet Skill Demand Shocks from CHIPS and Science
Co-Authors: Christophe Combemale & Krishnan Ramayya
Abstract:
Expansionary industrial policies, such as the CHIPS and Science Act, are followed with notable surges in labor demand within the industries they target. In the case of the CHIPS and Science Act, industries such as semiconductor manufacturing experienced significant influxes in financial investment, with $231 billion being committed to Semiconductors & Electronics thus far. Recognizing the imperative to address such labor demand shocks, we propose a novel operational methodology. This methodology, informed by economics, assesses potential supply-demand skill discrepancies, and incorporates factors such as the intertemporal occupational rates of transition and regional wage distributions. By analyzing the skill compositions inherent to industry-related occupations, our approach provides a strategic advantage to policymakers and industry stakeholders, enabling them to identify specific U.S. locales with the requisite skill profiles and potential wage structures. Furthermore, the practical application of our methodology is embodied in the Workforce Insights Tool, which offers comprehensive labor insights. To substantiate the efficacy of our approach, we consider the semiconductor manufacturing industry in the context of the CHIPS and Science Act as a representative case study, exploring diverse strategies for the construction of skill profiles for industry-related occupations.
A working draft is available at https://ssrn.com/abstract=4680624
Effect of COVID-19 Border Closures on Domestic Trade Patterns: A Spatial Autoregressive Approach to Estimate the Response of Colombian Trucking Flows
Abstract:
In this paper, I analyze and quantify the effects of exposure to international trade by considering the response in trade flow dynamics that took place within the Colombian domestic trucking network following the spread of COVID-19 and the implementation of spread-mitigation policies. Using transaction-level data on Colombian trucking for 2019 and 2020 along with geospatial trade exposure characteristics data, I implement a panel-data spatial autoregressive (SAR) model estimation to study how trade exposure affected the response of Colombian domestic trucking flows, measured as the value of goods traded between municipalities, to COVID-19 spread mitigation policies. Overall, the results suggest that accounting for spatial autocorrelations is important when conducting trade flow analyses and that the trading role of the municipality prior to COVID-19 did influence how a municipality was affected by COVID-19 and the spread-mitigating policies.
I made the poster below as part of my MS GIS-T capstone submission, and was developed from the paper presented above. This paper is still in-progress.
Estimating the Role of Trade Intermediaries in Colombia
Abstract:
In this paper, I analyze the role of intermediary firms, defined as firms that engage only in retail or wholesale capacities, in Colombia. Using Colombian transaction-level customs data for 2011 through 2016, I characterize the presence of intermediaries in international trade over time. I find that the distribution of the share of firms and share of values are relatively consistent over time. Gravity’s effect in trade is affirmed in the dynamics of firms’ exports; however, I identify a trend in imports that diverge from the literature, which has predominantly focused on developed countries for empirical estimation. This paper contributes to the literature by presenting a comprehensive overview of the types of firms conducting trade while considering a developing country.