Richard Norrie: How helpful is the government’s new measure of poverty?

The Government has announced that it will be publishing a new measure of poverty. In essence, it will be adopting the new statistical measure proposed by the Social Metrics Commission, to be published as ‘experimental statistics’ alongside the standard measurements of relative and absolute poverty that stem from the Households Below Average Incomes survey.

The adoption of this measure is one of the recommendations of United Nations Special Rapporteur for extreme poverty Philip Alston’s misguided and inaccurate report on poverty in the United Kingdom. Commenting on the adoption, Minister for Family Support, Housing and Child Maintenance Will Quince said:

“Our HBAI figures are National Statistics based on incomes and give us a strong statistically robust picture of the levels of poverty in the UK. However, the Social Metrics Commission makes a compelling case for why we should also look at poverty more broadly to give a more detailed picture of who is poor, their experience of poverty and their future chances of remaining in, or entering, poverty. We look forward to exploring the merits of developing a new measure with them and other experts in this field. In the long run this could help us target support more effectively.’

The standard measures of poverty are defined as follows: those in relative poverty have household incomes below 60 per cent of the median; those in absolute poverty have household incomes below 60 per cent of the 2010/11 median income. The Social Metrics Commission incorporates income as well as measures of available resources such as housing, childcare, and disability-related costs-of-living. A poverty threshold of 55 per cent below a three-year average of total resources available is set.

The SMC website boasts that it ‘unites poverty experts and thinkers from the left and right with a new poverty metric’ that ‘shows who is poor now and how that has changed over time, to provide a clear focus for policy makers’.

However, there is good reason to believe that this is untrue.

In statistics, the validity of a measurement — the extent to which it is measuring what it claims to measure — is often ascertained by comparing it to another conceptually related measure. This is known as criterion validation. One possible comparator is the Eurostat measure of severe material deprivation. Comparing this to the SMC poverty statistics gives us three reasons why the former might be an invalid measure of poverty.

(1) It is too high — the SMC figure of 22 per cent in poverty is much higher than the Eurostat share of people experiencing severe material deprivation — 5 per cent.

(2) The SMC poverty measure does not track the Eurostat severe deprivation index over time. As seen in the graph below, there is no obvious correspondence; indeed between 2009 and 2013, while the former is falling, the latter is rising.

Figure 1

 

(3) Changes in the SMC poverty measure are slight — between 2007 and 2009, it rises by 1.4 percentage points before falling again. Contrast this with the Eurostat deprivation measure more than doubling between 2008 and 2013.

A measure that finds that 22 per cent are in poverty when just 5 per cent experience deprivation will be including people who, while having less, want for very little of the essentials. Note that the dictionary definition of poverty is ‘the state of being extremely poor’. That the Social Metrics Commission poverty measure shows improvements at the same time that deprivation is more than doubling, makes a mockery of the SMC claims to providing an accurate measurement over time. It simply cannot be believed that poverty and deprivation move in different directions.

Perhaps it is the case that the SMC poverty measure is the valid one and the Eurostat deprivation index is the invalid one? Firstly, the Eurostat measure corresponds to data on subjective financial wellbeing from Understanding Society. And secondly, we have Ockham’s razor to contend with. This is the principle that, with all things being seemingly equal between two competing explanations, one plumps for the simplest, as it is more likely to be true. The Eurostat measure is compiled from a survey which classifies people as deprived if they cannot afford four out of nine common household items. Contrast this with the diagram below, taken from the SMC’s documentation, which explains how its measurement is derived. It is clear that the SMC measure is the more complicated.

Figure 2

The complexity of the SMC measure is another reason not to use it. At present, the standard measures of poverty are both easily understood and replicable by independent researchers. Government adoption of this measure makes it more difficult and time consuming for others to check the data and conduct their own research. This makes government less accountable.

Individual liberty is guaranteed by government and the rule of law. However, government can also be a threat to liberty; much like Baby Bear’s porridge, it is about finding the right balance. But a government that looks to create policy based on inaccurate statistics will be one liable to misfire, wasting taxpayer’s money and leading to interventions that are impossible to evaluate successfully and that do not meet the genuine needs of its citizens.

This does not mean to say that the standard measures of poverty should be maintained instead. My own analysis of Understanding Society data shows that while 13 per cent are in relative poverty (before housing costs), the majority of them describe their financial situation as at least ‘getting by’ if not better — see Table 1. Indeed, there were more people classified as in poverty, following standard conventions, describing their financial situation as ‘living comfortably’ than ‘very difficult’. I furthermore found that 41 per cent of those classed as in poverty had access to satellite television at home, which implies a disposable income spent on non-essential items. This is hardly consistent with most understandings of ‘poverty’.

Were the SMC measure of poverty to be submitted to a similar analysis, then I suspect the same patterns would emerge.

Table 1. Subjective financial situation by poverty status — Understanding Society Wave IV

Very difficult Quite difficult Getting by Doing alright Living comfortably
Poverty 10% 13.7% 34% 27.1% 15.1%
Not in poverty 2.3% 5.8% 23.4% 36.3% 32.2%
Overall 3.3% 6.8% 24.8% 35.1% 30%

Thomas Sowell has said that poverty statistics usually are just based on arbitrary thresholds, dreamt up by statisticians that due to that very arbitrariness, fail to capture the economic circumstances of real people. He jokes that they only really help combat poverty in the sense that they keep government statisticians out of poverty. It is right that we review how we measure poverty and seek accurate statistical measurement. However, the SMC approach is not it and it is disheartening that the government has chosen to adopt its measurement. Perhaps a better approach would be to use the Eurostat measure of severe deprivation: it gives a more realistic appraisal and has a much better track record of measuring genuine economic hardship over time.

Dr Richard Norrie — Researcher, Integration Hub and All in Britain

Leave a Reply

Your email address will not be published. Required fields are marked *