Plastic bag ban, causal policy analysis using NLP and citizen data

Jakarta banned retailers from distributing single-use plastic bags in 2020. Policymakers needed evidence that the ban worked before extending similar rules to other waste. However, no earlier quantitative study had measured its effect using the city’s own complaint platforms.

The ban began during COVID-19 lockdowns, which also reduced shopping and waste. Therefore, a simple before-and-after comparison would mix these effects. I needed to control for mobility, economic activity, seasonal waste, and changes in complaint use.

I used difference-in-differences to compare changes in plastic complaints with changes in other waste complaints. I classified 100,000+ records by waste type using text classification. The records came from JAKI, Jakarta’s official smart city complaint platform, and Qlue, a civic reporting platform. The classes separated plastic complaints from general waste, sanitation, and other urban services. Plastic complaints formed the treatment group. Meanwhile, non-plastic waste complaints formed the control group within the same platform and period.

I controlled for COVID-19 mobility using Google Community Mobility Reports. Moreover, I controlled for seasonal waste patterns and changes in platform use. This separated changes in waste from changes in reporting behaviour.

I found a statistically significant reduction in plastic complaints after the ban, controlling for pandemic effects. Moreover, I presented the pre-registered hypotheses and transparent method at ICISS 2021.