Their stats are diddled with biases and not applicable to the US. Their author.

Our study has several limitations that open up room for future research. First, we have a survivorship bias in our data since we are not able to identify those social startups that failed during our observation time. Further studies could therefore expand the literature on social startup survival and examine whether social startup trademarking also reduces the likelihood of failure. Second, our results might suffer from omitted variable bias. Since prior research shows that social startups that receive early external funding have higher employment growth rates and greater economic performance (e.g., Cacciolatti et al., 2020; Rey-Martí et al., 2016), this might be a variable to explain some of our trademark effects. For example, it could be the case that trademarking increases the likelihood of external funding, which then increases employment growth rates. Therefore, we encourage future research to investigate this mediation relationship since we were not able to collect funding rounds of social startups within their first three years of existence. Furthermore, future studies could use further proxies for the economic and sustainability outcomes of social startups. While the number of created jobs is an established growth parameter (e.g., Henderson, 2002; Lall and Park, 2020), research thus far lacks quantitative evidence on the sustainability outcomes of social startups. Since our proxy of sustainability outcomes of a social startup is an indirect measure, we encourage further research to replicate our findings based on a more direct measure of sustainability outcomes (e.g., using a survey approach). Such a study would allow a more detailed distinction between different SDGs and their achievements. From the perspective of trademark research, such a study would help to clarify the underlying mechanisms of the relationship between IPRs and SDGs. Third, our findings relate to German social startups, which reduces the generalizability of our findings since sustainable entrepreneurship has been proven to differ geographically (e.g., Defourny and Nyssens, 2010; Fichter and Tiemann, 2018; Kerlin, 2006). Future research should thus investigate other geographical areas, in particular, since the use of IPRs also varies between different regulatory contexts (e.g., Cohen et al., 2002; Maskus, 2000).