In this article, the authors extend the results of Aguirre-Urreta, Rönkkö, and Marakas (2016) concerning the omission of a relevant causal indicator by testing the validity of the assumption that causal indicators are entirely superfluous to the measurement model and discuss the implications for measurement theory. Contrary to common wisdom (Hardin & Marcoulides, 2011; Jarvis, MacKenzie, & Podsakoff, 2003; Kim, Shin, & Grover, 2010; McCoach & Kenny, 2014), Aguirre-Urreta et al. (2016) showed that the exclusion of one of the relevant formative indicators does not bias structural parameters and does not alter the empirical meaning of the focal latent variable. Aguirre-Urreta et al. (2016) join Howell (2014); Lee, Cadogan, and Chamberlain (2013) and Rhemtulla, Bork, and Borsboom (2015), who claim that the formative side of the focal latent variable is superfluous. This would mean that only the reflective side (outcomes) of the focal latent variable is of interest, the formative side of the latent variable does not add further information and is entirely superfluous to the meaning. The analysis by Aguirre-Urreta et al. (2016) clearly showed that a relevant causal indicator without negative consequences on the structural parameters can safely be excluded. However, the Monte Carlo simulation examples used by these authors did not enable the conclusion that the formative side of the focal latent variable is superfluous as claimed by Rhemtulla et al. (2015). This article extends the work of these authors and adds further results clarifying the validity of the assumption that causal indicators are superfluous.