This book is concerned with one of the central questions in social policy: what difference does difference make? My main concern in this respect is the difference that places make to people's health. The purpose of this chapter is to introduce the book's framework for exploring this question, set out the policy background, and consider how places matter for health and offer settings for interventions.

People's health in the UK has shown a pattern of steady improvement over several decades. Between 1971 and 2003, life expectancy for males increased from 69.1 years to 76.2 years, and for females from 75.3 to 80.5 years (ONS, 2005a). By 2003, however, there was a gap in estimated life expectancy between professional and unskilled manual social classes of over eight years for males and almost five years for females. Between the best and worst local authority areas the gap was nearly 12 years. Men and women living in the most deprived areas not only live shorter lives than those in more affluent areas, they spend twice as many years in poor health.

It is sometimes argued that the role of local action to tackle health and other social problems is at best marginal to, and at worst a distraction from, the real issues of income and wealth inequality that are the underlying causes of these problems (Pantazis and Gordon, 2000; Ball and Maginn, 2004). The same criticism was made of area-based initiatives (ABIs) to tackle deprivation in the 1970s, a very different era when income inequalities were much narrower than they are now (Community Development Project, 1977). Is it that we have not learned from the 1970s or do places still matter?

Places matter because they are open, dynamic and adaptive systems that do not have a simple cause—effect relationship with national or global drivers of economic, social or policy change. No strategy for tackling health inequalities will reach everyone it should without intervention in neighbourhoods to tackle the local factors that combine with wider determinants of health to create preventable geographical inequalities. This is because there are processes of local emergence at work. These can be investigated empirically but, as with all empirical work, we need a theoretical guide to where to look, what to look for, and what to make of what we see.