Figure 5.
The figure has two stacked panels showing event study estimates with points and vertical confidence intervals from relative time negative 5 to positive 5, with a dashed horizontal line at 0 and a vertical reference at time 0. Panel a shows cash acres minus homesteaded acres normalized. Estimates stay close to 0 before time 0, then rise sharply after, reaching about 0.006 at time 2, dipping near 0.003 at time 3, and increasing to about 0.0065 by time 5. Panel b shows cash patents minus homestead patents. Pre treatment estimates fluctuate around 0. After time 1, values increase strongly, reaching about 23 at time 2, falling near 16 at time 3, rising again near 23 at time 4, and ending around 17 at time 5.The effect of PA adoption on cash sales vs. homesteads
Note(s): This figure presents the event study plots corresponding to our preferred specification in column 5 of Table 3, which presents estimates of the impact of statutory recognition of PA on differences in land settlement via cash sales vs. homesteads using the estimator developed by De Chaisemartin and d’Haultfoeuille (2020) that is robust to dynamic treatment effects. Panel (a) presents results for normalized patented acres, and panel (b) presents results for total patent counts. Event study plots for other specifications from Table 2 can be found in Figures A4 and A5. (a) Cash acres – homesteaded acres (Normalized); and (b) Cash Patents – homestead patents