There is a strong suggestion from the existing literature that volunteering improves the wellbeing of those who give up their time to help others, but much of it is correlational and not causal. In this paper, we estimate the wellbeing benefits from volunteering for England's National Health Service (NHS) Volunteer Responders programme, which was set up in response to the Covid-19 pandemic. Using a sample of over 9,000 volunteers, we exploit the oversubscription of the programme and the random assignment of volunteering tasks to estimate causal wellbeing returns, across multiple counterfactuals. We find that active volunteers report significantly higher life satisfaction, feelings of worthwhileness, social connectedness, and belonging to their local communities. A social welfare analysis shows that the benefits of the programme were at least 140 times greater than its costs. Our findings advance our understanding of the ways in which pro-social behaviours can improve personal wellbeing as well as social welfare.