Funferal

Measuring poverty in Ireland – relative and absolute measures

Bernie Goldbach has a piece on his blog about poverty – or what he sees as the lack of it – in Ireland, citing my recent response to Friedman as one of his sources at the end.
There seem to be three main planks to his argument. First, commonly quoted statistics about the high numbers living in Ireland are misleading because there is no way that 22% of those living in Ireland are impoverished. Second, we are all poor relative to someone else. Third, Ireland has good social supports. All of which leads to a conclusion that Ireland has its poverty but also offers more safety nets than many other countries.
I take issue with Goldbach’s argument and conclusion. At heart I think the problem is not any lack of understanding or mistakes on his part but rather an ideological difference over the purpose of measurements of poverty. Goldbach believes that the statistic quoted should be of some absolute measure of what he terms impoverishment. In contrast, as I’ve discussed here previously, the trend in Europe in particular is towards more holistic measures such as social exclusion. The strength of such measures is that they reveal the extent to which the benefit of increased prosperity is being shared across society – the extent to which we have a fairer society in which those across the socio-economic spectrum see improvements as the gross measures of economic health (GNP per capita, etc.) improve.
In fact, the Irish government seems, to my eye, to be somewhat behind the trend in relying on figures for relative poverty. Further, the government has increasingly quoted rates for consistent poverty in order to mask the problems that relative poverty statistics reveal. As Labour spokesperson on social and family affairs, Willy Penrose, noted last September:

The Government continues to use the measure despite the fact that it is no longer relevant to actual poverty as experienced by people up and down the country. The much more widely used and internationally accepted definition of poverty – that is, any household with an income of less than half the average household income – is not referred to by the Government because it is inconvenient.

Penrose also, helpfully, provides some statistics on poverty in Ireland from 1997 to 2001:

So we can see why quoting rates of consistent poverty would suit the government much better than the rate for the 50% poverty line. (Note, by the way, that the figure for the 50% poverty line is, of course, a more conservative measure of poverty than, for example, using the 60% poverty line.) Obviously, these figures can be seen to support one part of Goldbach’s thesis, that when we use some measure of ‘absolute’ poverty the rate is much lower than the relative poverty rates.
But let’s look at just what the ‘absolute’ measures cover. Penrose points to concerns that the use of consistent poverty as a measure is inappropriate:

By 2002, the ESRI, the originator of the measure, was pointing out that there were potentially serious problems in continuing to use this 15 year old measure without major revision. By 2003 the NESC had added its concerned to the continuing use of the ‘consistent poverty’ measure stating that the credibility and usefulness of monitoring poverty targets will be undermined if the camera remains trained on what may no longer move.

At issue is the fact that the measure, developed for the Irish situation by the ESRI in 1987, no longer provides a meaningful measure of deprivation. Originally designed to provide a more comprehensive picture of poverty than merely looking at income levels,

The consistent poverty measure, as well as looking at income levels, also added 8 specific indicators of deprivation such as; unable to afford a warm waterproof overcoat or no substantial meal in the past two weeks and so on.

The eight measures had been chosen after research that involved talking with those living in poverty about what aspects of their deprivation had most effect on their lives. The problem is that as culture and social setting change so do the items that cause one to feel, to be, excluded from greater social life. A similar problem is faced in measuring items such as inflation in the consumer price index, where the typical basket of goods used to produce the overall rate can become increasingly irrelevant to the way people actually live their lives (which is why the basket used is revised once every decade or so).
Goldbach and others may, of course, argue that rather than try to make the measure more reflective of actual life in Ireland we should move towards a more universal, objective measure. Two problems arise here, however. First, the cost of living differs across countries. One would need to adjust our measure for relative costs. In effect we would need to create a measure based on Purchasing Power Parities (PPPs). Unfortunately, the standard PPP figures used by the OECD and others wouldn’t be sufficient for our purposes here, as they are, as with consumer price indices, based on an average basket of goods that would not be representative of the spending patterns of those living in poverty.
Second, and more importantly, at what level would you set the bar? On a global basis the UN and others have focused attention on the roughly 21% of the population who live on $1 a day, what they term extreme poverty. Would this be a reasonable level? What about the $2 a day level at which half the world live? Would it then be meaningful in any way to say that someone in Ireland with an income of $2.50 a day – $912.50 a year – was not in poverty? Obviously, part of the reason why the answer has to be no is based on the relative costs referred to above.
Additionally, however, what one needs to survive and thrive is in many ways socially determined. In the United States, for example, in many areas having access to a car is necessary if one wants to be able merely to travel to work. Obviously, greater availability of public transport alleviates this need so this is not merely an income-driven question.
We can also look at things like Maslow’s Hierarchy of Needs. At what point in his hierarchy would we set the ‘impoverished’ bar? Would we say that it’s enough that people have food and a roof to no longer be in impoverished? What about the example (in the linked page) of a woman suffering from an abusive husband – would we feel that someone should have sufficient material resources to be able to escape such a situation? What about the roughly 50 million Americans who do not have healthcare insurance – something which is necessary if one is to be able to have access to any minimally-acceptable standard of healthcare in that country, as discussed here recently. Healthcare insurance (together with professionalization of the providers and many other factors) has shaped the nature of the healthcare available in the United States, to a point where alternative modes of delivery which might be available in more impoverished countries are not actually acceptable or available in the United States (or if they are are often niche markets catering to affluent market segments).
Setting aside the particulars of Maslow’s theory – with which I am not personally over-enamoured – and linking to the point about socially-determined needs, what would it mean in Ireland today to not have the resources to be able to access any of the telephone, television and radio, books and newspapers? So much of Irish life today is centred on these mediated communications that to not have access is to be excluded from social, political and cultural life – it becomes difficult to get a job if one cannot provide a phone number at which one can be contacted or if one cannot (in these days when payphones are being increasingly phased out) access a telephone to respond to advertisements.
Much of Irish social entertainment used to be based around music sessions in homes. Nowadays sessions take place in pubs, so one needs to be able to afford to attend a pub if one wants to have access to what was, pre-commodification, a basic level of entertainment. One could, of course, buck societal trends and hold a session of one’s own – if one has access to musical instruments. So, do we include the ability to purchase musical instruments, or the ability to occasionally enjoy a pint, in our objective measure? If not, how is it that an abstract measure would now leave more people culturally impoverished, but not recognised by the model? One response to Goldbach’s post implied that mostpoor households in Ireland have 2-3 televisions. What this snide remark misses – even if accurate – is that television is actually a very cost effective means of entertainment. Even with a television license fee, and including electrical costs, a television provides virtually unlimited entertainment for as little as €4 a week. With the cheap satellite dishes now being marketed, which provide access to unencrypted satellite television, this is one of the most cost-effective investments a household can make – less than the cost of a daily newspaper – with access to news, entertainment and the other benefits of television. So if people object to them having television, are they saying that there really isn’t a problem until people have no access to any form of entertainment and are completely alienated from the political and current affairs of the day? Somehow I think that approach just stores up trouble for you.
To be fair to Goldbach he does note that divisions in society do not equate to an impoverished society. The problem comes in assuming, therefore, that the most useful and important statistic to quote would be one relating to absolute impoverishment – a measurement I think I’ve shown is problematic in itself. To take a topic Goldbach addresses occasionally on his own blog, education. Goldbach is opposed to the free third-level education we have in Ireland, another point on which I am at odds with him, but we can leave that specific argument to one side for the moment. Goldbach has detailed the benefits in terms of typical earning power of receiving a third-level education in Ireland – something I’ve also covered recently. Incidentally, the (2003) report Goldbach used indicated that the benefit of third-level education (in terms of increased income) was greater in Ireland than in other European countries, while my analysis of a recent study indicated the opposite, though the research I examined didn’t separate out number of years of third-level education so that may be a factor (especially if an unusually large proportion of those with third-level education only have a diploma).
But I digress. Whatever figures one uses, it’s obvious that access to third-level education has a large impact on future earning potential. It’s also a fact (insofar as we have them in these postmodern times) that the proportion of the population who attend third-level is very low amongst those from less-privileged socio-economic backgrounds. Goldbach may disagree but I’m convinced that the manner in which social opportunities – for whatever complex matrix of reasons – with such far-reaching implications are being fore-closed from large proportions of our population based on their inherited economic circumstances is serious, and it is for reasons such as this that I believe that measures such as rates of social exclusion are the proper measures to examine.
Update: When writing this earlier I was unable to locate a link to a recent story I know I’d seen about decreasing numbers of payphones. I’ve now located the text of that story which originated in last Saturday’s Irish Times. The main point of that story is that:

Details of Eircom’s payphone rationalisation strategy are contained in a new filing to the Securities and Exchange Commission (SEC). It shows that on March 31st, 2005, Eircom operated a network of 6,050 payphones across the Republic. This compared to 6,300 in March 2004.

Note that the story appeared in the business pages. This supports the claim often made by political analysts that developments like this are often framed primarily as business stories, in terms of their impact on profit margins, and only on a secondary level in terms of their impact on the marginalized members in society.
In fairness to the Irish Times they do mention the reasons why reductions in the number of payphones are controversial, but remember that in the print edition of the newspaper this is physically in the midst of stock reports and similar items, where it is less likely to be accessed by those who do not consider themselves insiders in business:

Eircom’s payphone rationalisation programme, which began in 2001, has saved the company millions of euros in operating costs by reducing the number of staff required to maintain the boxes and collect coins.
But it has in the past proved controversial among rural communities, some of which have complained about plans to remove payphones in their areas.
Eircom must provide payphones throughout the Republic under the terms of its universal service obligation, a key regulation mandated by ComReg.