The commonly agreed indicators for social inclusion that we described in Chapter 3 are in general outcome indicators aiming to measure the extent of progress towards the common objectives of promoting social inclusion. To bring about a substantial improvement in the reported indicators requires long-term and structural policy efforts in the fields of economic growth, social protection, minimum wages and employment. It has, moreover, to be recognised that the outcomes measured by the indicators depend partially on developments outside the control of Governments (such as trends in family formation and dissolution). This is the reason why most Member States highlighted in their NAPs/inclusion lists of policy measures and policy-related indicators, which can be more easily integrated within the development of a policy strategy, such as the number of unemployed or long-term unemployed persons who are assisted by some labour market measure, the number of available social housing units and the amount of minimum income benefits. What we need to do, however, is to link policy and outcomes. Crucial to the EU Social Inclusion Process is a better understanding of this relationship. The purpose of this Chapter is to consider how this analysis can be strengthened. What are the key elements in establishing the relationship between policy measures, as listed in the NAPs/inclusion and their EU analyses by the Commission and Member States (Joint Reports on Social Inclusion, Joint Reports on Social Protection and Social Inclusion and Commission Staff Working Papers), and outcomes, as measured by the commonly agreed indicators? What are the strong and weak points of different types of analysis? How can they be applied at Member State and EU level?

We begin in Section 4.1 with a brief survey of the policy analysis presented in the NAPs/inclusion prepared in 2003 by EU-15 countries and in 2004 by the new Member States as well as the EU analysis of these national reports in Joint Report and Commission Report (European Commission, 2004b and 2004d), identifying some of the directions in which it could be developed. For this purpose, it is necessary to make use of modern tools of policy analysis, and in Section 4.2 we examine two types of approach: model families analysis and micro-simulation models. The ground covered will be familiar to many readers, but, as stressed in Chapter 1, we would like to make the book accessible to those who have not been engaged in the technical debates.