Javier Milei's surprising victory in Argentina's 2023 presidential election unfolded amid pronounced economic turmoil, with inflation at 138% and poverty affecting 42% of the population. We introduce a high-frequency digital-trace approach to assess whether real-time economic anxiety contributed to Milei's rise. Using Google Trends as a proxy for public attention, we construct a continuous daily time series of candidate-related searches using a novel normalization strategy that overcomes Google Trends' scaling limitations. We integrate this data with daily informal exchange rate data and official inflation statistics to test competing hypotheses about which economic signals drive political salience. Using Bayesian seemingly unrelated regression (SUR) models, we show that currency shocks captured by the informal "Dólar Blue rate," and official annual inflation are associated with increases in candidate-related searches. Our analysis illustrates how digital trace data can complement traditional approaches to studying election-related dynamics.
Levin et al. (Sat,) studied this question.