Voice and Exit: Mutual Funds’ Reactions to ESG Scandals

Do fund managers push for change after controversies, or do they quietly sell and move on?

Fatima Zahra Filali-Adib, Daniel Schmidt, and Bastian von Beschwitz study how mutual funds respond to ESG controversies in their paper « Voice and Exit: Mutual Funds’ Reactions to ESG Scandals ».

They explore whether funds use voice (voting) or exit (divestment), and how a fund’s own controversy experience shapes these choices. Their main conclusions include:

• ESG scandals are rare but price-relevant: RepRisk index jumps above 25 occur in 0.58% of stock-months, yielding 3,213 scandals in the dataset used for this study.
• Around controversies, cumulative market-adjusted returns from 5 days prior to 5 days after are -0.79%, and abnormal trading volume rises 22.9%.
• Scandal exposure is associated with outflows: -11.8% in fund flows, with -12.4% in active funds and smaller and statistically insignificant changes in passive ones.
• On voice, a 1% increase in ESG controversies raises the probability of voting in favour of an ESG shareholder proposal by 1.4%, a 4% increase relative to the mean.
• The estimated effect is stronger when the controversy is tied to an above-median position, when it represents a « large surprise », and when it is accompanied by below-median stock returns.
• On exit, a 1% increase in ESG scandal experience raises the probability of selling a high-ESG risk stock by about 0.72%, with little variations up to three quarters in the future.

This article shows flows, voting, and divestment are jointly determined, which suggests stewardship KPIs should be evaluated alongside commercial pressure and portfolio rebalancing.

Fund mangers exit positions with high sustainability risks even when this reduces voting power, rather than scaling holdings to maximise engagement impact.

Proxy votes are used as a measurable engagement proxy, but private engagement channels such as meetings or letters are not in the data, which could understate « voice » activities happening off the ballot.

The authors also caution that ESG scandal classifications are fluid and can evolve over time, making it hard to map E, S, and G categories cleanly to economic channels.