The objective of this article is to analyze invalid votes under compulsory voting in Peru. I analyzed three Peruvian presidential elections (held in 2006, 2011, and 2016). I investigated the results of these elections at the municipal level, examining the first and second rounds separately. Thus, I created an original dataset that includes 5,478 cases from each round. I used the ordinary least squares (OLS) regression models to test theories related to invalid votes. First, I found that the higher the level of development in municipalities, the lower the number of invalid votes in municipalities. Second, I examined a feature of runoff voting and found that the higher the share of votes for eliminated candidates in the first round, the greater the number of invalid votes in the second round. These results suggest that invalid voting is linked to socioeconomic grievances and that electoral abstention under non-compulsory voting transforms into invalid votes in the second round under compulsory voting when voters’ choices are limited. These findings contribute to the current research on invalid voting, electoral systems, and electoral behavior in Peruvian municipalities.