Social capital is a theoretically confusing concept, but one which nonetheless has much to offer as a potential measure of the strength of societies. It is in need of theoretical examination. This paper will outline the major theoretical aspects of social capital in relation to its three main component parts – values (such as trust), the networks such values are relevant to, and the practices (such as volunteering) related to those values and networks. It will also outline the major problems with the concept identified in the literature so far, including conceptual issues such as tautology and outcomes, the issue of who owns social capital, the possibility of different types or ‘axes’ of social capital – bonding and bridging – crowding each other out, and the existence of negative social capital (which incorporates issues of ‘victim blaming’). In addition, two more uncommon problems specific to measurement are identified. The first is the predominance of analysing only the micro-individual demographic influences upon social capital – or the bias towards agency. Social capital is seen as shaped by individuals through their personal characteristics. However, large macro-social forces shape social capital as well. Such forces are more structural in nature, and are embedded within material reality. Four macro-social forces stand out in particular – materialism, inequality, gender-family dynamics, and cultural clashes – and their impacts are in need of measurement. This leads onto the final problem with social capital; choosing the appropriate level at which to analyse trends. International comparison to identify ‘emergent properties’ and policies peculiar to countries or clusters of countries that influence social capital is needed, particularly in relation to the influence of macro-social forces. Analysing clusters of countries by welfare regime is suggested here to these ends. In the final section of the paper, hypothetical connections are drawn between established social capital concepts and welfare regimes. Suggestions for empirical research are presented in conclusion.