Background: Social capital can be conceptualised as an individual resource residing in relationships between individuals or as a collective resource produced through interactions in neighbourhoods, communities or societies. Previous studies suggest that social capital is, in general, good for health. However, there is a shortage of studies analysing the association between individual and collective social capital in relation to health amongst older people. Purpose: The purpose of this study was to assess the relationship between municipal- and individual-level social capital and self-rated health amongst older people in Western Finland and Northern Sweden. Method: Data were retrieved from a cross-sectional postal questionnaire survey conducted in 2010. The study included, in total, 6,838 people aged 65, 70, 75 and 80 years living in the two Bothnia regions, Västerbotten, Sweden and Pohjanmaa, Finland. The association between social capital and self-rated health was tested through multi-level logistic regression analyses with ecometric tests. Social capital was measured by two survey items: interpersonal trust and social participation. Results: Individual-level social capital including social participation and trust was significantly associated with self-rated health. A negative association was found between municipal-level trust and health. However, almost all variation in self-rated health resided on the individual level. Conclusions: We conclude that contextual-level social capital on a municipal level is less important for understanding the influence of social capital on health in the Bothnia region of Finland and Sweden. On the other hand, our study shows that individual-level social participation and trust have a positive and significant association with self-rated health. We suggest that other ways of defining social capital at the collective level, such as the inclusion of neighbourhood social capital, could be one direction for future research.