Working Papers
“The Rise of Broadband and the Retail Apocalypse: Evidence from Consumer Packaged Goods” (Job Market Paper) R&R at the Journal of Marketing
I explore consumer behavior in the US consumer packaged goods (CPG) sector during broadband’s proliferation from 2006 to 2016. Using household panel and retail scanner data covering over 40,000 brands across 900 categories, I analyze nine household and retailer outcomes: brands purchased, trips taken, retailers visited, offline spending, any online spending, online spending share, prices, price dispersion, and demand elasticities. I combine US Census and FCC data to track the roll out of broadband. Contrary to “retail apocalypse” fears, aggregate CPG trends are muted. Exploiting geographic variation in broadband growth, I find economically small but heterogeneous effects. Digging deeper reveals offsetting generational behaviors, with declines in brick-and-mortar shopping among younger consumers counterbalanced by stability in larger older cohorts, and a concerning “hollowing out” of the middle class. I demonstrate robustness using terrain instruments, within-household adoption, and statistical power tests. This study provides the most comprehensive examination to date of broadband’s impact on CPG shopping. The results suggest broadband is driving a gradual evolution rather than a dramatic upheaval in CPGs, while surfacing challenges in engaging key consumer segments.
“Employment Impacts of Implementing an Employer’s Contributory Pillar: Evidence from Chile,” with Marcela Parada-Contzen, Lucas Provoste, Cristobal Sanhueza, and James Traina R&R at Journal of Pension Economics and Finance
his paper estimates the labor demand price elasticity in Chile using five waves of the Chilean
Longitudinal Survey (2007-2017). The estimates are used to simulate the impacts on employment of implementing an employer’s contributory pillar in the pension system. We estimate specific price elasticities for eleven industries and four worker categories (i.e., managers, clerks, skilled and unskilled production workers). We find that labor demand is negative and inelastic for all specifications. Our estimates are consistent with previous findings. In the baseline model, price elasticity ranges between 0.304-0.911% as wages increase by 1%. We predict increases in unemployment rates between 0.20 pp and 0.71 pp from a baseline unemployment rate of 6.51% as wages increase by 1%. Job losses are predicted in the range of 17,789 to 63,937 positions. Not considering heterogeneity in elasticities by industry and type of worker causes the underestimation of the impact of increasing labor costs by 0.51 pp. Demand price elasticity and job losses are higher for unskilled workers.
“Several Million Demand Elasticities,” with Jordan Rosenthal-Kay and James Traina
The household’s price elasticity of demand is a key input to many economic models’ construction of markups and the assessment of consumer surplus. We measure the price elasticity of demand for around 14,000 products by region-year using retail scanner data. In all, we estimate over 7.5 million demand elasticities. We find that the distribution of these elasticities is stationary over time. However, we document substantial spatial heterogeneity in consumers’ price sensitivity: consumers in the largest markets are the most price elastic. As demand elasticities are a key input into the measurement of markups, our results suggest that any conclusions that markups are rising in retail markets must be driven by assumptions on conduct.
“Does Information Mitigate Behavioral Gaps Due to Market Inexperience” with Forrest Spence
We collect detailed individual level purchasing and search data on 2,449 consumers in the UNC-Chapel Hill textbook market. A random subset of these consumers were provided with information about the price of their assigned textbook from various retailers before the semester began. We use these data to compare the search and purchasing outcomes of consumers who received information from their instructors (treatment group) with outcomes of consumers who did not receive this information (control group). We find that (i) information leads to more online search and online purchases, (ii) the informational treatment reduces the gap in online search and purchasing behavior between consumers across experience levels, and (iii) the treatment increases take-up of the textbook at the extensive margin.
Works in Progress
“Information Technology Effects on Advertising” with Ningyin Xu
We use variation in the growth of broadband to study how the rise of information technology has shaped advertising. From 2010 to 2019, broadband rates increased from 70% to 80%. At the same time, TV advertising remained steady while spending as a proportion of advertising has decreased in other channels. Our results show that a 10 percentage point increase in broadband increases spending in offline advertising in that DMA by approximately 1.2%. Similarly, a 10 percentage point increase in broadband increases spending in TV advertising in that DMA by approximately 1.1%. We show that broadband decreases print advertising in most DMAs.
“The Rise of Private Label” with Guenter Hitsch
We highlight two findings: a convergence in household PL share purchases and divergence in retailer channel PL share sales. Our first finding is that households have converged in their PL purchases. That is, households with the lowest initial PL consumption in 2004 saw the sharpest rise from 2004-2018. Our second finding is that channels have diverged in their PL sales. We decompose the overall aggregate trend by retailer channel to reveal that private labels purchases are not growing uniformly across retailer types.
Publications
Balaban, Rita, Donna Gilleskie, and Uyen Tran. “A quantitative evaluation of the flipped classroom in a large lecture principles of economics course.” The Journal of Economic Education 47, no. 4 (2016): 269-287. DOI: 10.1080/00220485.2016.1213679