Recent studies on the effects of anti-smoking policies on subjective well-being present mixed results and do not account for potential externalities, especially among couples. We contribute to the literature by evaluating the impact of smoking bans on well-being externalities among smokers and non-smokers as well as couples of different types of smokers. We exploit the policy experiment provided by the timing of the UK public smoking bans and measure well-being via the GHQ. We employ matching techniques combined with flexible difference-in-differences fixed effects panel data models on data from the British Household Panel Survey. The joint use of matching methods with fixed effects specifications allows building more comparable treatment and control groups, producing less model-dependent results and accounting for individual-level unobserved heterogeneity. We find that public smoking bans appear to have a statistically significant short-term positive impact on the well-being of married individuals, especially among women with dependent children. These effects appear to be robust to alternative specifications and placebo tests and are discussed in the light of the economic theory and recent evidence.