Why This Question Matters
Karl-Oskar Lindgren, PƤr Nyman, and Sven Oskarsson test whether a simple text-message reminder can change voting behavior and whether those effects spread to others. Understanding these spillovers matters for scholars and practitioners who design low-cost mobilization campaigns: if reminders not only affect recipients but also household members or coworkers, the reach and cost-effectiveness of such interventions change considerably.
A Massive Randomized Field Test
The authors report results from a pre-registered, well-powered field experiment of text-message get-out-the-vote (GOTV) reminders during the 2019 European Parliament election in Sweden (N > 3,000,000). Recipients were randomly assigned to receive a short SMS reminder to vote. Outcomes were compared for treated individuals and for people linked to them in two social contexts: household members and workplace colleagues.
Key Findings
- The direct causal effect of receiving a text message was small but measurable: an increase in turnout of 0.3 percentage points.
- Roughly half of that direct effect appears to spill over to untreated household members, while spillovers among coworkers were essentially zero.
- Heterogeneous effects: the direct treatment effect is noticeably larger for individuals with below-average baseline voting propensity; however, within that low-propensity group, household spillovers were significantly negative.
Interpreting the Surprising Negative Spillovers
The authors explore why some household spillovers are negative for low-propensity groups. They present indirect evidence that treated individuals are more likely to vote early, and argue this timing change may weaken the usual mechanisms that generate positive spilloversāspecifically, early voters are less likely to bring family members to the polling place, reducing opportunities to mobilize household contacts. The negative spillovers are presented as a speculative but plausible account, supported by suggestive patterns in the data rather than definitive proof.
What This Means for Mobilization and Research Design
These results show that simple SMS reminders produce small direct increases in turnout but can have meaningful secondary effects within households that vary by voter type and timing of voting. For campaigners, the findings imply that the net reach of reminders depends on who is targeted and when recipients vote. For scholars, the study highlights the importance of measuring spillovers and considering voting timing (early vs. Election Day) when evaluating field interventions.
