Legislation Effective Against Hate Speech on the Internet

Hateful content is widespread on social media. Almost without restraint, users disregard rules and criminal laws on the internet and insult, defame or threaten other people. To counteract this, the Network Enforcement Act (NetzDG) was enacted in Germany in 2017. The regulation is effective in curbing harmful content on the internet, as a recent study by ZEW Mannheim shows.

Using the short message provider Twitter as an example, the researchers prove that both the number and the intensity of hateful comments have measurably decreased due to the law. At the same time, this leads to fewer hateful comments being retweeted or liked – so the law has also contributed to a somewhat friendlier atmosphere on the internet. The ZEW study shows that the number of hateful tweets has been reduced by an average of 10 per cent as a result of the legislation. However, to capture the full effect of the regulation, the number of reactions from the user community must also be taken into account. In the study, hateful messages were retweeted an average of four times. Thus, if a hateful content is not shared, the usual chain of further hate tweets is interrupted, which contributes to a more appropriate use of the platform.

“The study shows the higher effectiveness of legal regulation to curb harmful content on the internet compared to simply relying on the self-commitment of social network providers. This is particularly important given that hate speech is increasingly followed by deeds and can even lead to violent riots like the one at the US Capitol a year ago. Comparable laws based on the German model are currently being discussed both in the EU and in other countries. In order to choose the most effective measures, the effects should be analysed in great detail,” says Raphaela Andres, researcher in ZEW’s “Digital Economy” department and author of the study.

“One possibility for further curbing hateful content might be to simplify the complaints procedure for users. Currently, they have to name the exact offence themselves and sign that a false accusation is a violation of the platform regulations. Reporting hateful messages thus becomes a risk for the user and an obstacle to a more efficient fight against hate on the internet,” says Olga Slivko from Erasmus University Rotterdam and co-author of the study.