Introduction

In this chapter, we put evidence from earlier chapters within a common framework, and give an overview of what this shows about the impact of policies towards poverty, inequality and social exclusion under New Labour.

A first danger is timing. If the 1950s were still “too early to tell” the impact of the French Revolution according to Zhou Enlai, 2004 is far too early fully to assess policies that are still being implemented. The problem is not just of present preoccupations, but also of data. Statistics follow events with a lag. While there has been great improvement in the speed of key poverty and income distribution statistics, the most recent available to us are for 2002/03, with a mid-point in autumn 2002, before the major April 2003 tax credit changes, for instance. Other statistics have longer lags. More fundamentally, many measures are designed to have long-term effects: the impacts on adult outcomes of new policies towards children's early years and later education are inevitably still unknown.

Second, by focusing on specifics, we may miss the bigger picture. We have tried to avoid concentrating only on the government's own agenda, but there is often more to say when policy has been active than when it has not. Many data are from government sources, and relate to its own targets and priorities. It is impressive that there is now an annual progress report on poverty and social exclusion, Opportunities for all (DWP, 2003d) – and that it could report in 2003 that 33 of its 43 indicators had improved over the medium term (generally since New Labour came to power) while ten were steady and none had deteriorated. However, external views may be more convincing. For instance, the New Policy Institute's collection of indicators over a similar period shows 21 of its 44 indicators as improving, 16 steady, but seven deteriorating (Palmer et al, 2003). This is less rosy than the official collection but nonetheless encouraging, and contrasts with its earlier assessment of ‘New Labour's inheritance’, when they had found 19 indicators improving in the medium term, 11 steady, but 14 to have been deteriorating (Howarth et al, 1998).