A century of inequality in England and Wales using standardized geographical units

This paper is concerned with very long-terms trends in poverty and inequality in England and Wales over the last century. It seeks, in particular, to provide a tentative quantitative answer to the question of whether relative poverty has become more or less extreme geographically (Dorling and Woodward 1996). Such a study is only possible through the application of novel methods for data analysis based around the Great Britain Historical Geographical Information System (GBHGIS) (Gregory and Southall 1998), which we use to make data for different dates as closely comparable as possible. The range of variables available for use is inevitably limited because we require statistics that were measured for detailed geographical areas over the course of the century, and which remained both significant and broadly comparable over the whole period. Even so, we believe that both the methodology and the results are sufficiently interesting to justify this paper and its findings. If the three very different indicators of inequality used here all show the same trend towards increased geographical inequality, then it is likely that poverty as a whole is becoming more geographically unequal. Our analysis is strictly quantitative, using data for the whole of England and Wales interpolated onto a set of standardized spatial areas. This permits us not only to clearly define the geography of relative inequality in different periods, but also to make some broad generalizations concerning long-term trends in relative inequality. In other words, we are concerned with establishing which geographical areas experienced the worst hardship, how much worse off they were relative to the best-off areas, and how this pattern changed over time. Owing to the limitations of early data the areas used are relatively large and crude, being approximately equivalent in scope to modern Travel-to-Work areas. This puts us at the top of a multi-scale approach to the study of poverty and inequality (Shepherd 1998) and should complement more detailed and localized studies. The debate about how best to define poverty is as old as the study of poverty itself and many different methods have been put forward (see for example Gordon and Pantazis 1997; Lee 1998). Our time scale restricts us to three key indicators, each of which is discussed in more detail in the subsequent sections. Each can be associated with one of William Beveridge’s five great evils, identified by him at the mid-point of our study: