The number of mail ballots cast in a given election has increased over time, indicating a shift in voter behavior with respect to the method of voting. Traditionally associated with older, white, educated, and affluent voters, mail voting has also been linked to low levels of voter confidence. However, a nuanced relationship exists, as individuals might opt for mail voting if they are confident their vote will count, or their confidence might increase after casting a mail ballot and becoming familiar with the process. Therefore, the presence of simultaneity is the primary source of bias when estimating the causal effect of voter confidence on mail voting. To control for this source of potential bias, I employ a robust set of instruments and a two-stage least squares approach to identify the effect of voter confidence on mail voting. Utilizing data from the 2016-2024 Survey of the Performance of American Elections (SPAE) and an instrumental variable approach, findings show that those with higher levels of voter confidence, overall, are more likely to cast a mail ballot.