Rediscovering budget standards methods

In the early 1980s I became interested in expenditure data (Godfrey and Bradshaw, 1983). At the time I was directing the Social Policy Research Unit (SPRU) at the University of York and some of us in the unit were doing research on the impact of rapidly rising energy costs following the Yom Kippur War (Bradshaw and Harris, 1983). We were also doing work on social security (Deacon and Bradshaw, 1983) and were aware that the Thatcher government was introducing policies which would eventually lead to a more than doubling of child poverty during the decade. I was active in the Child Poverty Action Group (CPAG), chaired at the time by Peter Townsend. The Thatcher government had just replaced the Low-Income Statistics (LIS) series on poverty, based on the social assistance scales, with what would become the Households below Average Income series – still with us today – based on (now but not then) a 60 per cent median income threshold. This data was not having any impact on policy. A measure based on the proportion of households with incomes below 60 per cent of median equivalent income did not seem to resonate or move the public, although poverty was increasing rapidly.

We were in York, the City of Seebohm Rowntree, who had completed three very influential studies of poverty in the city based on budget standards published in 1901, 1941 and 1951 (Rowntree, 1901; Rowntree, 1941; Rowntree and Lavers, 1951). Rowntree advised Beveridge (1942) and the second survey results were available to him including the budget standard but, as Veit-Wilson (1992) has pointed out, the national insurance rates recommended by Beveridge were based on less eligibility with minimum subsistence assumptions. A budget standard was a specified basket of goods and services which, when priced, can represent a standard of living. Rowntree's budgets were designed to establish an income to meet basic minimum necessities. Beveridge's benefits aimed at minimum subsistence. But budgets could be devised to represent any living standard.

Budget standards methods had disappeared from living standards research in the UK in the postwar period (though David Piachaud (1979) published his Cost of a Child based on budget standards).

Introduction

The objective of this chapter is to operationalise and test out a number of possible measures of extreme poverty applied to the European Union countries, using the analysis of household survey data from EU Statistics on Income and Living Conditions (SILC). It is a revision and updating of two similar analyses undertaken by Bradshaw (2000 and 2001) and Bradshaw and Mayhew (2011), both undertaken for the European Commission. That work had been commissioned because of a dissatisfaction felt by some elements in the Commission with the measures of poverty that were being conventionally used in the European Union. So we start by reviewing developments in the conventional measurement of poverty in the EU. Then the concepts of absolute and extreme poverty are discussed. The bulk of the chapter presents the results of an analysis of the poverty rates, poverty gaps and poverty composition of five conceivable measures of extreme poverty.

These are:

• • The World Bank dollar per day concept.

• • Poverty thresholds based on national minimum income schemes – social assistance.

• • A threshold set on the basis of a minimal reference budget standard.

• • A severe deprivation standard.

• • An overlaps measure based on severe deprivation and low income. The latter is the preferred option.

Conventional measures of poverty

Following the Lisbon European Council in 2000, the European Union was committed to fight against poverty and social exclusion using the Open Method of Coordination (OMC). A key element of the OMC was a set of indicators agreed upon jointly by the European Commission and all EU member states, in order to measure progress towards the agreed EU social inclusion objectives. At the 2001 Laeken European Council, 18 indicators were adopted.

Subsequently, the Social Protection Committee (SPC) and its Indicators Sub-Group (ISG) developed these. The main measure of monetary poverty included in the EU list of indicators is a relative one (net income less than 60% national median), known as the ‘at-risk-of-poverty’ rate. Since the EU Council of Ministers meeting in 1975, poverty in the EU has been conceived of as relative to a particular country at a particular time. There was and is strong justification for this approach rooted in social science understandings.

Introduction

This chapter details findings on child poverty and social exclusion from the 2012 UK Poverty and Social Exclusion Survey (PSE-UK 2012). The 1999 Poverty and Social Exclusion Study of Great Britain (PSE-GB 1999), which at the time offered the most comprehensive data on child poverty and social exclusion gathered in Britain, was completed amidst an atmosphere of hope in relation to the outlook for poor children and families. New Labour’s landslide victory in the 1997 general election was followed by then Prime Minister Tony Blair’s commitment in 1999 to end child poverty within a generation. This commitment was made in the aftermath of 18 years of Conservative governments, under Prime Ministers Margaret Thatcher (1979–90) and John Major (1990–7), who had overseen drastic increases in the child poverty rate. Lloyd (2006), in her analysis of child poverty in the PSE-GB 1999, details many of the policy changes and interventions designed to combat child poverty. These changes culminated in the Child Poverty Act 2010, which committed the government to monitoring and reporting on progress in relation to four targets to be achieved by 2020:

• • fewer than 10% of children in relative poverty (equivalised household income <60% national median, before housing costs);

• • fewer than 5% of children in combined low income and material deprivation (equivalised household income <70% national median before housing costs and having a score of 25 or less on the Households Below Average Income child material deprivation measure; see Carr et al, 2014 for more details);

• • fewer than 5% of children in absolute poverty (equivalised household income <60% national median of the base year [2010/11], fixed in real terms);

• • fewer than 7% children in persistent poverty (equivalised household income <60% national median for three out of the previous four years; target set October 2014).

Child poverty declined steadily over the course of the Labour government, and the Child Poverty Act 2010 passed through Parliament with cross-party support. However, long before that milestone, analyses of the progress being made towards policy goals demonstrated that greater efforts would be needed to achieve the targets.

Introduction

Previous chapters in this volume explore different dimensions of poverty and social exclusion in the UK, while the companion volume (Dermott and Main, 2017) presents evidence on the poverty and social exclusion experienced by different social groups or in different locations. Although many of these contributions also examine how disadvantage in one area relates to disadvantage in others, they essentially focus on one or two aspects of exclusion at a time. This chapter seeks to bring these analyses together by presenting an overarching multidimensional analysis of social exclusion, exploring in more depth the relationships between the dimensions of exclusion. As noted in the Introduction, there are many ways in which this task could have been approached. Rather than selecting just one, this chapter explores complementary approaches using a range of statistical techniques. By doing so, we hope the results provide a more holistic understanding of how different domains of exclusion relate to each other and of the different forms which social exclusion may take.

The chapter is organised into a number of sections. The first provides an introduction to the concept of social exclusion, while the second describes levels of exclusion using the main indicators available in the PSE-UK 2012 survey and, where available, how these have changed since 1999. The chapter then discusses two different approaches to constructing broader measures of exclusion for each domain or groups of domains, in order to facilitate the multidimensional analysis which follows. The results of this multidimensional analysis are then presented, exploring the relationships between domains, and the variations in the forms of exclusion across different social groups, including between poor and non-poor groups. Lastly, the chapter considers the wider significance of these findings for policies to promote social inclusion and well-being in the UK.

The concept of social exclusion

As noted in the Introduction, the concept of social exclusion emerged into UK discourses on poverty and living standards from France in the early 1990s (Bradshaw, 2004). Early advocates, including Room (1995), argued that it expanded on income- and expenditure-based measures of poverty by adopting a more dynamic, multidimensional perspective which emphasised the structural processes underpinning disadvantage. Despite fundamental differences in views on its causes (and appropriate policy responses), some consensus is evident in its conceptualisation in academic and policy research.

Introduction

Two decades of transition have seen growing diversity across Eastern Europe and Central Asia, both in terms of economic and institutional structures and social developments. Living conditions have changed rapidly during this period, bringing both new opportunities and at the same time greater vulnerabilities for families and children. Economic recovery during the past decade has been hit again by the global economic crisis – many countries in the region belong to the most severely affected countries in the world. While most governments have tried to protect social spending, measures to mitigate the impacts of the crisis are only as good as existing social protection systems. However, these too often prove to be ineffective in reducing child poverty or supporting families with children, not least because cash transfers are too low and too narrowly targeted at the extremely poor to make much of a difference in the lives of poor people. In view of crisis responses, the narrow targeting of cash transfers also reduces the flexibility and responsiveness needed to address new vulnerabilities (Hoelscher et al, 2009).

Rapidly changing living conditions require monitoring systems that are sensitive to qualitative changes in the lives of children. Currently the most important regional monitoring tool is the TransMONEE database, tracking child-related administrative data across Central and Eastern Europe (CEE) and Central Asia (see, for instance, UNICEF, 2009). At country level, on the other hand, much of the monitoring of child well-being has followed the Millennium Development Goal (MDG) indicators which focus mainly on basic health (for example, mortality and immunisation) and education indicators (for example, enrolment in primary education). While these indicators are important, they fail to capture changes in the quality of life of children and their families in transition countries. These countries traditionally have had fairly well developed healthcare and education systems and basic access was often maintained for the majority of the population during the transition period. The bottlenecks are somewhere else: in education, for instance, problems are not so much found in access to primary education but in preschool and secondary education as well as in the declining quality of education.

In the UK we have no official means of establishing how our children are doing, no ‘State of UK children’ report. This book is an attempt to fill that gap. It is the fifth in a series of volumes that has been produced out of the University of York stable since Bradshaw (2001). That book emerged from the Economic and Social Research Council (ESRC) Children 5-16 research programme, and was motivated by anxiety about the impact that the doubling of relative child poverty rates under the Thatcher government might be having on the well-being of children in the UK. It concluded by recommending that the UK should produce a regular review of the well-being of its children, and that perhaps responsibility for it could be taken by the Office for National Statistics (ONS). When this looked unlikely, Save the Children (UK) stepped in and supported the publication of two further volumes – Bradshaw (2002) and Bradshaw and Mayhew (2005). The last volume was produced with the support of the Child and Maternity Public Health Observatory (CHIMAT) in Bradshaw (2011).

The 2011 volume reported the situation of children around 2009/10, and focused on the years of the Labour government during 1997-2010. It really covered a period before the global financial crisis in 2008, and certainly before the impact of the austerity measures introduced by the coalition government (elected in 2010) could be felt. Time has passed on, and another cohort of children has gone through infancy, primary and secondary schools, and much that might affect children has changed since the last volume.

Up until 2008 the UK had experienced an unprecedented period of economic growth, with employment rates reaching record levels, and unemployment had never been so low. During this period, there were substantial increases in public expenditure on education, health, transport and cash and tax benefits for families with children. Other significant developments for children were the establishment of Sure Start, the extension of nursery education to three- and four-year-olds and substantial investment in pre-school childcare.