Radical
Statistics
Success measures: response to the government Green Paper on welfare reform
David Gordon and Peter Townsend
This paper is a written memorandum on the proposed success measures outlined in the Welfare Reform Green Paper New Ambitions for our Country: A New Contract for Welfare (Cmd 3805). In particular, it deals with the DSS HBAI series and the relevant tables in Social Trends.
The need to measure poverty and social exclusion
2 One of the most important tasks of the Welfare State and the DSS is to alleviate poverty and combat social exclusion. This is a view that is shared by both Ministers and the general public alike. Very few people wish to live in a country with a high endemic level of poverty - it is unjust, expensive and socially divisive.
3 In order to produce cost-effective and efficient ways of eradicating poverty, it is necessary to define and measure poverty accurately in the first place. It is impossible to tell if a policy is working well unless you have some idea of what would constitute success. Unfortunately, for the past 20 years the British Government has refused to accept or adopt any definition of poverty and has made little attempt to measure poverty in the UK.
International definitions of poverty
4 Although the Government has refused to accept a definition of poverty for use in the UK, it has, over the past 25 years, signed various international treaties and agreements which have incorporated definitions of poverty. For example, in 1975, the Council of Europe adopted a relative definition of poverty as: "individuals or families whose resources are so small as to exclude them from the minimum acceptable way of life of the Member State in which they live" (EEC, 1981). The concept of ‘resources’ was defined as: "good, cash income, plus services from public and private resources" (EEC, 1981).
5 On the 19 December 1984, the European Commission extended the definition as:
"the poor shall be taken to mean persons, families and groups of persons whose resources (material, cultural and social) are so limited as to exclude them from the minimum acceptable way of life in the Member State in which they live" (EEC, 1985).
6 After the World Summit on Social Development in Copenhagen in 1995, 117 countries (including the UK) adopted a declaration and programme of action which included commitments to eradicate "absolute" and reduce "overall" poverty, drawing up national poverty-alleviation plans as a priority (UN, 1995).
7 Absolute poverty was defined by the UN as "a condition characterised by severe deprivation of basic human needs, including food, safe drinking water, sanitation facilities, health, shelter, education and information. It depends not only on income but also on access to services" (UN, 1995, p. 57).
8 Overall poverty was considered to take various forms, including "lack of income and productive resources to ensure sustainable livelihoods; hunger and malnutrition; ill health; limited or lack of access to education and other basic services; increased morbidity and mortality from illness; homelessness and inadequate housing; unsafe environments and social discrimination and exclusion. It is also characterised by lack of participation in decision-making and in civil, social and cultural life. It occurs in all countries: as mass poverty in many developing countries, pockets of poverty amid wealth in developed countries, loss of livelihoods as a result of economic recession, sudden poverty as a result of disaster or conflict, the poverty of low-wage workers, and the utter destitution of people who fall outside family support systems, social institutions and safety nets." (UN, ibid, p.57).
9 Thus the UK Government has signed international agreements that include broad definitions of both absolute and relative poverty and has agreed to actively implement policies to reduce the extent of poverty and exclusion in the country. Unfortunately, none of the currently available Government-sponsored surveys or statistical series is specifically designed to measure poverty or social exclusion. It is, therefore, difficult to see how success in reducing poverty and exclusion will be either monitored or achieved.
The current low income statistics
10 During the last 20 years, there has been no official national survey of poverty in Britain. Those concerned with poverty and inequality at a national level have had to rely almost exclusively on the Low Income Families series and its successor the Households Below Average Income (HBAI) series, produced by the DSS and the Social Services Select Committee. These have generated data on the size and characteristics of the population living on a low income over time and have proved useful in revealing the extent of and growth of financial inequality. Without them we would not be able to say very much about trends in inequality since the early 1970s. However, they are limited because they are based exclusively on the Family Expenditure Survey (and now the Family Resources Survey) and have been restricted to a threshold of relatively low income - such as the lowest decile and half average household income. There has been no sustained comparative analysis of high and low income households.
The problems with financial measures of poverty
11 Financial measures of poverty - most commonly income but also expenditure - have been employed, despite the fact that, since Townsend (1979), it has been widely recognised that income is an inadequate measure of command over resources. Measurement of income fails to take account of personal and capital assets, fringe benefits and occupational welfare, income in kind, the value of free or subsidised services and the quality of the environment. Despite the hundreds of academic articles written on poverty and the lively debates surrounding issues of poverty measurement, it is now universally accepted that poverty is about more than just low income - it is also concerned with the standard of living that people have. Low income is not problematic if it is planned for and/or short lived. For example, many self employed people expect and make adequate allowance for, low profits when first establishing a business. It is the combination of low income and low standard of living that constitutes poverty.
12 The use of income as an indirect measure of poverty has also involved arcane debates about the appropriateness of the equivalence scales used to adjust income to need, the problems of measuring the incomes of the self employed, whether income should be assessed before or after housing costs, what is the most appropriate income threshold, whether it should be related to mean or median incomes and whether poverty numbers, poverty gaps or some combination of both should be measured (see Townsend and Gordon, 1989, 1992; Townsend, 1996 for discussion). Indeed, these dilemmas or choices have been shown to make a significant difference to the estimate derived of the size and structure of the poor population. For example, despite using a common equivalence scale, the HBAI 1994/95 estimates of the proportion of the population (individuals) living in households in poverty varies from 6% to 32% depending on whether 40, 50 or 60% of average income is used as the threshold, whether the self employed are included or excluded and whether income is measured before or after housing costs. Indeed, the definition of income poverty has determined not just the size but the composition of the poor population and this, of course, has an impact on the appropriateness of the policy response to poverty (Bradshaw et al, 1998).
13 Politicians have been able to take refuge from the clear evidence of rapidly increasing rates of poverty behind this debate about definitions. Thus, for example, one year, commenting on the publication of the HBAI figures, Peter Lilley claimed that the income poverty figures were an overestimate because the expenditure poverty figures produced lower estimates. The next year he claimed that the poverty estimates were exaggerated on the grounds that analysis of income poverty over time showed that there was a good deal of turnover of the poor population and this was therefore a reason not to be concerned about the overall level of poverty (Hills, 1998). In fact he misinterpreted and exaggerated the limits of the data but this episode illustrates how the authority of estimates of poverty based on purely financial indicators have become undermined (Bradshaw et al, 1998).
How to measure poverty
14 Poverty is a widely used and understood concept but its definition is highly contested. The term ‘poverty’ can be considered to have a cluster of different overlapping meanings depending on what subject area or discourse is being examined (Gordon and Spicker, 1998). For example, poverty, like evolution or health, is both a scientific and a moral concept. Many of the problems of measuring poverty arise because the moral and scientific concepts are often confused. In scientific terms, a person or household in Britain is ‘poor’ when they have both a low standard of living and a low income. They are 'not poor' if they have a low income and a reasonable standard of living or if they have a low standard of living but a high income. Both low income and low standard of living can only be accurately measured relative to the norms of the person’s or household’s society.
15 A low standard of living is often measured by using a deprivation index (high deprivation equals a low standard of living) or by consumption expenditure (low consumption expenditure equals a low standard of living). Of these two methods, deprivation indices are more accurate since consumption expenditure is often only measured over a brief period and is obviously not independent of available income. Deprivation indices are broader measures because they reflect different aspects of living standards, including personal, physical and mental conditions, local and environmental facilities, social activities and customs. Figure 1 below illustrates these concepts:
Figure 1: Definition of Poverty
16 The ‘objective’ poverty line/threshold is shown in Figure 1. It can be defined as the point that maximises the differences between the two groups (‘poor’ and ‘not poor’) and minimises the differences within these two groups. For scientific purposes broad measures of both income and standard of living are desirable. Standard of living includes both the material and social conditions in which people live and their participation in the economic, social, cultural and political life of the country.
Meaningful success measures of standard of living
- It is clear that, in order to achieve adequate success measures for monitoring the effectiveness of anti-poverty and anti-exclusion policy, it is essential to have meaningful measures of both low income and deprivation (low standard of living). The most widely used method of measuring deprivation is the consensual approach first developed by Joanna Mack and Stuart Lansley (1985) and recently updated by Bradshaw et al (1998). This is a simple and straightforward method of gaining information on people’s standards of living, using just eight questions and four showcards. It could be incorporated as a question module into virtually any of the current Government-sponsored social surveys but should ideally be included in the Family Resources Survey.
18 This approach has been tried and tested in both Britain and other countries and been shown to produce reliable and repeatable results. Nationally representative surveys in Britain have been undertaken by MORI in 1983 and 1990 (Mack and Lansley, 1985; Gordon and Pantazis, 1997) and in Wales in 1995 by Beaufort Research (Gordon, 1995). The Office of Population, Censuses and Surveys (OPCS) used a similar set of questions to measure the standard of living of disabled adults and families with disabled children in Britain in 1985 (Martin and White, 1988; Smyth and Robus, 1989). Similarly, representative surveys were carried out by the Policy Planning and Research Unit (PPRU) amongst disabled people in Northern Ireland in 1990 and 1991 (Zarb and Maher, 1997). The European Statistical Office (Eurostat) has used a similar set of questions to measure standard of living in Britain and the 14 other member states annually since 1994 as part of the European Community Household Panel Survey (Ramprakash, 1994; Vogel, 1997). This approach to measuring standard of living has also been adopted in Denmark (Mack and Lansley, 1985), Sweden (Halleröd, 1994, 1995, 1998), Ireland (Nolan and Whelan, 1996), Belgium (Van den Bosch, 1998), Holland (Muffels et al, 1990), Finland (Kangas and Ritakillio, 1998), Germany (Andreß and Lipsmeir, 1995) and Vietnam (Davies and Smith, 1998)
19 By directly measuring standard of living, it will be possible to identify the numbers of people suffering from material deprivation, social deprivation and social exclusion. Those who are excluded because of financial constraints, lack of services or lack of other resources can be distinguished from those who exclude themselves out of choice. A good success measure needs to be able to distinguish constraint from choice if cost efficient policies to alleviate poverty and social exclusion are to be developed and monitored.
Meaningful success measures of low income
- The main problem with the HBAI series is that the adjustments used to compare different-sized households (equivalisation) and the low income thresholds used (40, 50 or 60% of average income) are essentially meaningless. It is impossible to tell from the HBAI whether or not a household with an income below these income thresholds has sufficient money to live decently or not. It would be much more meaningful to produce low income statistics which show how many households do not have an adequate income to allow them to participate in the economic, social, cultural and political life of British society and avoid poverty. Low income thresholds and statistics should measure adequacy not arbitrary thresholds. The most widely used method of achieving this is to use Budget Standards approach.
- The Budget Standards approach can be used to produce estimates of the adequacy of income using the Family Expenditure Survey or similar surveys. This approach avoids most of the pitfalls of just using crude equivalised income figures as a proxy for poverty as in the current HBAI series. A budget standard is a specified basket of goods and services which, when priced, can represent a particular standard of living. Budgets can be devised to represent any living standard (Bradshaw, 1993).
22 Budget standards are probably the oldest method of exploring living standards, pioneered by Rowntree (1901) in his famous studies of poverty in York. They were used by Beveridge (1942) in the setting of the National Assistance scales, and then they went into the doldrums and have not been used much in Britain since the Second World War (with the exception of Piachaud’s study in the late 1970s). A budget standard estimates what families ought to spend rather than what they actually do (or think they need to) spend. A characteristic of this approach is that the judgements of ‘experts’ are used to create a basket of goods and services which represents the type of commodities, quantities and quality of family consumption (Oldfield and Yu, 1993).
23 Producing budget standards is expensive and extremely time consuming since it requires a team of ‘experts’ to first decide on the contents of the budget (housing, transport, food, clothing, etc.) and for this budget to then be priced for different types of household in different circumstances. Ideally, the British Government should undertake the necessary research to produce its own budget standard as the Australian Government is currently in the process of doing (Saunders, 1996). However, if the funds are not available, then the Low Cost but Acceptable budget standard produced by the Family Budget Unit at York University (Parker, 1998) would provide a meaningful basis for a low income statistical series.
24 A single budget standard or set of budget standards should be used to set the low income thresholds in the HBAI series instead of the current arbitrary thresholds of 40, 50 and 60% below average income. The low income statistics would then relate to a meaningful and readily understandable standard of living which allowed for the costs of being able to participate in the normal activities and customs most people take for granted. These statistics would then be a useful success measure since they would help reveal how effectively and rapidly welfare policy is working.
Conclusion
25 If the Government wishes to tackle the problems of social exclusion and poverty effectively and efficiently, then it requires valid and reliable success measures. The currently available HBAI statistical series is inadequate for this purpose. The proposed indictors (lone parents and disabled people in touch with the labour market) are useful but not sufficient for measuring the success of policies designed to alleviate the multifaceted problems of poverty and social exclusion.
26 Two sets of statistics are required to effectively measure the success of anti-poverty and exclusion policy. Firstly, the Government needs low income statistics that relate to a known, measurable and socially approved standard of living. These could be obtained via a modified HBAI series by abandoning arbitrary equivalisation methods and income thresholds and, instead, using a budget standard approach.
- Secondly, the standard of living that people actually have needs to be directly measured. The consensual method (Mack and Lansley, 1985) would reveal how many people currently have too few resources to allow them to escape from material and social deprivation and to participate in economic, social, cultural and political life of Britain. It would also distinguish between those who are excluded by their own choice and actions from those who are socially excluded due to a lack of income, services and other resources.
REFERENCES
Andreß, H.J. (Ed.) (1998), Empirical Poverty Research in a Comparative Perspective, Ashgate: Aldershot
Andreß, H.J. and Lipsmeir, G. (1995), Was gehört zum notwendigen Lebensstandard und wer kann ihn sich leisten? Ein neues Konzept zur Armutsmessung, Aus Politik und Zeitgeschichte, Beilage zur Wochenzeitung Das Parlament, B, 31-32/95 (28 July)
Beveridge, W. (1942), Cmd 6404, Social Insurance and Allied Services, HMSO: London
Bradshaw, J. (Ed.) (1993), Budget Standards for the United Kingdom, Avebury: Aldershot
Bradshaw, J., Mitchell, D. and Morgan, J. (1987), Evaluating adequacy: the potential of budget standards, Journal of Social Policy, 16(2), pp.165-82
Bradshaw, J., Middleton, S., Townsend, P. and Gordon, D. (1998), Perceptions of Poverty and Social Exclusion, Report to the Joseph Rowntree Foundation: Bristol
Callan, T., Nolan, B. and Whelan, C.T. (1993), Resources, Deprivation and the Measurement of Poverty, Journal of Social Policy, 22(2), pp.141-172
Davies, R. and Smith, W. (1998), The Basic Necessities Survey: The Experience of Action Aid Vietnam, Action Aid: London
EEC (1981), Final Report from the Commission to the Council on the First Programme of Pilot Schemes and Studies to Combat Poverty, Commission of the European Communities: Brussels
EEC (1985), On Specific Community Action to Combat Poverty (Council Decision of 19 December 1984), 85/8/EEC, Official Journal of the EEC, 2/24
Gordon, D. and Pantazis, C. (Eds.) (1997), Breadline Britain in the 1990s, Ashgate: Aldershot
Gordon D. and Spicker, P. (Eds.) (1998), The International Poverty Glossary. Zed Books: London, New York
Halleröd, B. (1994), Poverty in Sweden: A New Approach to Direct Measurement of Consensual Povert, Umeå Studies in Sociology, No. 10 6, University of Umeå
Halleröd, B. (1995), The Truly Poor: Indirect and Direct Measurement of Consensual Poverty in Sweden, Journal of European Social Policy, 5(2), pp.111-29
Halleröd, B. (1998) Poor Swedes, Poor Britons: A Comparative Analysis of Relative Deprivation, in Andreß, H.J. (Ed.), Empirical Poverty Research in a comparative Perspective, Ashgate: Aldershot
Hills, J. (1998), Income and Wealth: The Latest Evidence, Joseph Rowntree Foundation: York
Kangas, O. and Ritakallio, V.M. (1998), Different methods – different results? Approaches to multidimensional poverty, in Andreß, H.J. (Ed.), Empirical Poverty Research in a comparative Perspective, Ashgate: Aldershot
Mack, J. and Lansley, S. (1985), Poor Britain, George Allen & Unwin Ltd: London
Martin, J. and White, A. (1988), The Financial Circumstances of Adults Living in Private Households, Report 2, HMSO: London
Muffels, R., Kapteyn, A., Vries, A.de. and Berghman, J. (1990), Poverty in the Netherlands: Report on the Dutch Contribution to an International Comparative Study on Poverty and the Financial Efficacy of the Social Security System, VUGA: The Hague
Nolan, B. and Whelan, C.T. (1996), Resources, Deprivation and Poverty, Clarendon Press: Oxford
Oldfield, N. and Yu, A.C.S. (1993), The Cost of a Child: Living Standards for the 1990s, Child Poverty Action Group: London
Parker, H. (Ed.) (1998), Low Cost but Acceptable: A Minimum Standard for the UK, Policy Press: Bristol, (forthcoming)
Piachaud, D. (1979), The Cost of a Child: A Modern Minimum, Child Poverty Action Group: London
Ramprakash, D. (1994), Poverty in the Countries of the European Union: A Synthesis of Eurostat's Research on Poverty, Journal of European Social Policy, 4(2), pp.117-128
Rowntree, B.S. (1901), Poverty. A Study of Town Life, Macmillan: London
Saunders, P. (1996), Development of Indicative Budget Standards for Australia: Project Outline and Research Methods, BSU Working Paper No 1, University of New South Wales: Sydney
Smyth, M. and Robus, N. (1989), The Financial Circumstances of Families with Disabled Children Living in Private Households, OPCS Surveys of Disability Report 5, HMSO: London
Townsend, P. (1979), Poverty in the United Kingdom, London: Allen Lane and Penguin, and Berkeley University of California Press
Townsend, P. (1996), The Struggle for Independent Statistics on Poverty, in Levitas, R. & Guy, W. (Eds.), Interpreting Official Statistics, Routledge: London, New York
Townsend, P. and Gordon, D. (1989), Low Income Households, Memorandum of Evidence to the House of Commons Social Services Committee, 579, pp.45-73
Townsend, P. and Gordon, D. (1992), Unfinished Statistical Business on Low Income?: A Review of New Proposals by the Department of Social Security for the Production of Public Information on Poverty Statistical Monitoring Unit Report No. 3, University of Bristol
UN (1995), The Copenhagen Declaration and Programme of Action: World Summit for Social Development 6-12 March 1995, United Nations Department of Publications: New York
Van den Bosch, K. (1998), Perceptions of the minimum standard of living in Belgium: Is there a consensus?, in Andreß, H.J. (Ed.) Empirical Poverty Research in a comparative Perspective, Ashgate: Aldershot
Vogel, J. (1997), Living Conditions and Inequality in the European Union 1997, Eurostat Working Papers: Population and Social Conditions E/1997-3, Eurostat: Luxembourg
Zarb, G. and Maher, L. (1997), The Financial Circumstances of Disabled People in Northern Ireland, PPRU Surveys of Disability Report No. 6, NISRA: Belfast
David Gordon and Peter Townsend
Centre
for International Poverty Research
University
of Bristol
8
Priory Road
Bristol
BS8 1TZ
E-mail: Dave.Gordon@bristol.ac.uk Tel: (0117) 954761
E-mail: Peter.Townsend@bristol.ac.uk Tel: (0117) 954677
