Reviewed by Peter Saunders (CIS)

The author of this book is Director of the Social Policy Research Centre (SPRC) at the University of New South Wales. Although he shares the same name as me, we are not related in any way. Indeed, if this book is to be believed, we are bitter enemies in an ongoing ‘poverty war’.

The ‘war’ to which the book refers concerns the definition and measurement of poverty. One side in this war believes that poverty is a major problem in Australia, that it is getting worse under this government, and that some serious political intervention is needed if wrongs are to be righted. They want more generous welfare, higher taxes, extensive government job creation schemes and a lot more funding for academic research. According to this book, they are the good guys, and Saunders is one of them.

Against them stand the bad guys. They claim that the poverty statistics have been exaggerated, that the data are unreliable and that rich and poor alike have been gaining from the country’s rising prosperity. Saunders thinks these people are dangerous, and he warns that unless they are defeated, the poor will suffer and the ‘social fabric’ of the country will fray.[1] He is worried that the bad guys are winning, and he offers his book as the start of a fight-back.

Donning his homburg hat and sucking on a big cigar, Saunders tours the battlefield seeking to stiffen intellectual backbones using faux-Churchillian rhetoric not normally associated with a serious academic book. He reassures waverers that, ‘They can win [the war], and they should,’ and he rallies his troops by urging: ‘We need the courage to move forward and the determination to succeed.’[2] Trudging through craters left by the enemy’s empirical bombshells, and deftly avoiding the debris of his own side’s collapsed arguments, he laments the ‘lack of funding for research into the causes and consequences of poverty,’[3] and he sketches his plans for post-war reconstruction. He dreams of a ‘comprehensive anti-poverty strategy’ and ‘wide-ranging community consultation,’ and he foresees poverty experts like himself sitting at the heart of government as members of a ‘special unit…report[ing] directly to the Prime Minister.’[4] Committees fit for heroes.

Who are the protagonists in this ‘war’? Standing alongside Saunders on the side of the poor we find the usual rag-tag battalion of socialist academics, armed to the teeth with rhetoric and clichés, together with the Australian Council of Social Service (ACOSS), the research units of several prominent welfare organisations, the National Centre for Social and Economic Modelling (NATSEM), a smattering of trades unionists and church leaders, an assortment of Fairfax journalists and ABC commentators, a Labor-dominated Senate inquiry committee in charge of propaganda, and a bevy of welfare rights activists. Together, they make up the guys in white hats.

On the other side of the battlefield stands the Centre for Independent Studies (CIS). According to Saunders, they don’t care about the poor. Rather, they aim ‘to deny the existence of poverty and absolve government of responsibility for addressing it.’ Their arguments are ‘sinister’ and they fight dirty. They have ‘replaced reasoned argument’ with ‘abuse’. and they don’t respect the good guys (they have even had the gall to ‘challenge the professionalism and ethical standards of those who study poverty’).[5] Yet it’s these bandits in bandanas who are winning the poverty war (and I should declare an interest at this point and admit I am one of them).

The casual reader might think this is a bit of a lop-sided war, with hundreds of taxpayer-funded academics and welfare activists on one side and just the CIS on the other. The CIS operates on an annual budget that wouldn’t keep most university sociology departments in Guardian Weekly subscriptions and foreign conference trips for more than a few months. It rarely boasts more than four full-time-equivalent research staff at any one time, and frequently has to get by on two or three. Unlike the good guys, it receives no taxpayer funding to pay for its research (nor does it seek any, for every cent it spends is raised through voluntary membership fees, sales and donations). Yet despite (dare we say it?) its ‘relative poverty’. Saunders frets that the CIS has been winning all the battles, and he devotes a whole book to challenging its hegemony.

Like Richard III looking for his horse, Saunders expresses dismay at faint-hearted allies who have been turning and fleeing the battlefield. He complains that the Smith Family has ‘abandoned…its involvement in mainstream poverty research’ since having its research questioned by the CIS, and he moans that NATSEM ‘has been reluctant to re-enter the debate over the trends in poverty’ following CIS criticism of its methods. He thinks it is time to turn things around by ‘restoring credibility to Australian poverty research’.[6]  

One might think the best way to restore credibility would be to address and try to refute the most damaging claims that the CIS has been making. Mostly, however, this book avoids any such direct confrontation, preferring brief bursts of sniper fire to any sustained engagement on the core issues. Indeed, for a book claiming to chronicle a ‘war’. Saunders provides very little detail about the casus belli. Allow me, then, to fill in the gaps.

The ‘war’ in question dates back to 2001 when the Smith Family commissioned NATSEM to produce a report on poverty in Australia.[7] The report claimed that poverty was worsening despite economic growth and that more than 1 in 8 Australians was poor. It received extensive media coverage and attracted widespread assent from the academic community, yet its central claims were debatable. In the absence of critical scrutiny from anywhere else, the CIS took it upon itself to examine these claims in detail, and it found weaknesses in both the method of measuring poverty and the reliability of the income data.[8]

The main problem of measurement turned on the way the report defined ‘poverty’ as an income below half the mean income of the whole population. Any poverty line is arbitrary, but this one had some obvious problems of its own. If Bill Gates came to live in Australia, for example, his arrival would increase ‘poverty’ on this measure, for his income would push up the average income and thereby inflate the poverty line. Even more bizarrely, if the incomes of all those under the poverty line were raised to take them out of poverty, this would increase the mean income in the whole population, thereby raising the poverty line and reabsorbing many of these people back into ‘poverty’. Clearly there was something fishy about the way this poverty line was being drawn.

Even more disturbing than the method of measurement, however, were the anomalies in the income data the Smith Family report was using. The way we estimate what incomes people are receiving is on the basis of regular surveys conducted by the Australian Bureau of Statistics (ABS) in which samples of Australian households are asked how much money they have coming in. The obvious problem is that respondents may not report their true incomes accurately, and there are strong grounds for believing that under-reporting is particularly common among those households claiming to have the lowest incomes (the bottom 10% of the distribution, who make up most of those deemed ‘poor’).

A substantial proportion of these households deny they receive any income at all – no wages, no welfare benefits, no dividends or interest or rents or annuities, nothing. Some say they receive ‘negative incomes’ (i.e. they end up each week with less than they started with). And even those who say they get some money often give a figure well below what the welfare state would guarantee them if they bothered to step into any Centrelink office. In short, the reported incomes of those at the lower end of the ABS distribution are in many cases improbably low.

Nor do these reported incomes square with what people say they spend. The ABS has found that households in the bottom 10% of reported incomes say they are spending an average of 2.3 times more than they are receiving. Even allowing for the possibility of extensive use of credit cards, pawn shops and a good run on the TAB, this reinforces the suspicion that many of those identified as ‘poor’ are substantially under-reporting their true incomes. This is further borne out by the fact that the total value of welfare benefits imputed from the ABS income surveys falls well short of the total amount the Department of Family and Community Services in Canberra actually pays out to welfare claimants. Many of those at the bottom of the income distribution are living on welfare, but it is clear that significant numbers of them are neglecting to tell the ABS interviewers about all the benefits they actually receive.

When the CIS said all this in its critique of the Smith Family’s report, the social policy academics and welfare organisations reacted furiously. Many seemed indignant that their views and pronouncements should have been challenged – an impertinence they had not before encountered. UnitingCare said it was ‘absurd and offensive’ for the CIS to cast doubt on the poverty statistics. The St Vincent de Paul Society claimed it was obvious that there were ‘a lot of families and individuals living poor lives’. The President of ACOSS asserted that poverty is a major problem ‘whichever way you measure it’.[9] The basic message was that the CIS was being devious by pointing to problems in the way the research has been done and inaccuracies in the data. Poverty is widespread whatever the research shows, and it is irresponsible for CIS to query this judgement.

To its credit, the Smith Family did not try to shrug off or shout down the CIS criticisms. Instead, it approached Peter Saunders of the SPRC to prepare a response on its behalf. This he did, denouncing the CIS critique as erroneous and muddled, and reassuring the social policy community that there was nothing demonstrably wrong with the income data they had all been using. Imperiously he pronounced there to be ‘no reliable research that supports the claim’ that the income data were shaky, and he dismissed as ‘astounding’ the CIS suggestion that the ABS data were too unreliable to be used.[10] The tone and content of his comments implied that the CIS authors were amateur interlopers who had no hands-on experience of doing poverty research. His message was essentially that the experts should be allowed to get on with what they had been doing for the last 30 years, measuring poverty and warning the rest of us about how extensive it was becoming, and the upstarts from the CIS should shut up.

There matters might have rested had it not been for a most unexpected intervention. Just a few weeks after Saunders had issued his authoritative rebuttal of the CIS claims, an ABS Zeppelin appeared out of a clear blue sky and lobbed a WMD into the heart of his defences. The explosion came with an ABS announcement that the income data at the lower end were so unreliable that the Bureau itself would henceforth be ignoring the responses of the bottom 10% when analysing its own income distribution data. The Bureau went on to express its ‘concerns with the fact that the extremely low incomes (close to nil and sometimes negative) recorded for some households in this group do not accurately reflect their living standards’, and it warned that including data on the lowest 10% of recorded incomes in its analysis would ‘substantially lower the average income values’ and thereby give ‘a misleading impression of the economic wellbeing of the most disadvantaged households’.[11] 

The CIS, it transpired, had been right all along. The income data were too unreliable to be used, and far from being ‘astounding’, the CIS concerns were shared by the very organisation whose job it was to collect and publish these statistics. The ABS was telling the research community that the statistics were faulty and they should not be trusted.

This ABS bombshell did not end the ‘poverty war’, for the odd skirmish has continued to break out since, but every subsequent attempt by the poverty lobby to restore its earlier position has been undermined by the ABS fallout. In 2004, for example, a host of welfare groups and academics got an ALP-sponsored Senate poverty inquiry to agree that as many as 3.5 million Australians were living in poverty, but the Inquiry ignored all the ABS warnings about the problems in the income data, and its poverty estimate was worthless.[12] More recently, a few guerillas operating out of the St Vincent de Paul Society’s research bunker claimed that 4.5 million Australians are living in households receiving less than $400 per week, but again the estimate failed to adjust for all those people reporting impossibly low incomes, and the report nose-dived on take-off (in this case, to make matters worse, the researchers also got their math wrong).[13]

It is against this background that Saunders now seeks to re-open hostilities with a book intended to drive back the CIS and restore the poverty lobby to the high ground it once occupied. He admits that some of his allies urged him not to publish out of ‘fear that another book about measuring poverty will provide further ammunition for those all too keen to re-engage in another poverty wars skirmish.’[14] But he is undeterred by these faint hearts, believing the WMDs of the CIS have to be decommissioned if the good guys are ever to prevail.

So how does Saunders deal with the core issue at the heart of this ‘poverty war’ – the unreliability of the income statistics? Answer: he doesn’t. In a book of 155 pages, he devotes just two sentences to the whole issue. Recognising that it is ‘important to acknowledge’ some of the ‘technical problems’ that crop up in research on poverty, he goes on: ‘Aside from the problems determining what to include as income, people are reluctant to reveal their incomes in surveys. No less an authority than the ABS has raised concerns about the reliability of its own household income data, noting that those who report the lowest incomes often have higher expenditures, which suggests that income may understate the standard of living of those who, on the face of it, appear to be most susceptible to income poverty.’[15] And that is all he has to say about it. Nothing on whether the ABS data can still be used to calculate realistic estimates of the number of people in poverty; nothing on whether there is some other source of data that might support what the poverty lobby has been saying down the years about poverty being ‘extensive’, nothing on how the CIS claims about the unreliability of the income data might be countered. After more than three years, the fundamental issue has still not been addressed.

On the back cover of this book, Paul Smyth, who holds a chair at Melbourne University funded by the Brotherhood of St Laurence, asserts: ‘This book should end the poverty wars in Australia…it rescues brilliantly the study of poverty from the ideological sludge in which it has recently been mired.’ But how can a book like this ‘end the poverty wars’ when it runs away from the single, biggest issue that started the war and inflicted all the damage in the first place?

Saunders never attempts to defend the credibility of the income statistics. Instead, he seeks to prosecute his war on two other fronts. The first is to trash the reputation of the CIS (a strategy he also used back in 2002). The second is to shift the ground of the debate by broadening the definition of ‘poverty’ so as to minimise the importance of the income data. Neither strategy works.

We have already seen one example of how Saunders traduces the reputation of the CIS, accusing it of replacing reasoned argument with abuse (yet he offers no examples of where CIS has been abusive). In similar vein, he talks of a ‘barrage of assertions that defy logic and strain credulity’, but he offers no indication of how logic is defied or credulity strained in what the CIS has been writing.[16] He also attacks the ‘pro-market ideological predisposition’ of the CIS, but he never explains what is wrong with a pro-market stance (given that ever-increasing state spending has clearly failed, and that participation in the market economy is known to be the best route out of poverty, such a position might be considered a strength rather than a weakness).[17]

Running throughout the book is the repeated slur that, by criticising poverty research, CIS has shown its indifference to the situation of people who are poor – as if the poor might somehow benefit as a result of researchers exaggerating their numbers. The truth is that CIS has developed a clear policy strategy for combating poverty by getting people off welfare, opening up the labour market to generate more jobs, and reforming income tax to create stronger work incentives. For academics like Saunders, all three components of this strategy run against the grain, for they want higher welfare spending, more government regulation and increased taxes. Had his book attempted to address these core issues, it might have moved the debate forward in a constructive way. Instead, it draws the simplistic assumption that anybody who does not agree with the left’s prescriptions ipso facto does not care about the problem.[18]   

Saunders several times tries to represent the CIS as amateurish by pitting what it says against what various left-wing academic ‘experts’ say, the assumption being that the latter must be right. Defending the concept of ‘social exclusion’ against a CIS critique, for example, he contrasts what he calls the ‘sober but considered reflection’ of a favoured academic with the ‘wild claims and assertions’ of CIS researchers, but he never shows why the former should be believed over the latter.[19] Similarly, he attacks a CIS claim that a major international study, the Luxemburg Income Survey (LIS), exaggerates the extent of income inequality in Australia as compared with other countries. Saunders says the CIS is ‘way out of line in its understanding’ of this survey, but the only grounds offered for this haughty judgement are that ‘one of the world’s leading authorities on income distribution, Professor (Sir) Tony Atkinson’ thinks the survey is useful.[20] But does Atkinson’s knighthood and his academic title mean his opinions are automatically right and CIS is wrong?

The irony in both of these examples is that, having attacked the CIS for its ‘wild claims’ and its ‘out of line’ assertions, Saunders ends up agreeing with it! He ridicules the CIS for claiming that the Australian data in the LIS are not comparable with data collected for other countries, but he then makes precisely this point himself on the very next page.[21] Similarly, having decided there is ‘little justification’ for the CIS claim that the concept of ‘social exclusion’ could be applied to almost anyone, he goes on to admit that ‘defining social exclusion has proved to be an elusive task,’ and he quotes Amartya Sen as saying, ‘The language of exclusion is so versatile and adaptable that there may be a temptation to dress up every deprivation as a case of social exclusion.’[22] Precisely!

Most of the flak Saunders flings at the CIS thus misses its target, for it takes more than labels and name-calling to win an intellectual argument.

Saunders’s second, and more positive, strategy for reversing fortunes in the ‘poverty war’ involves sidestepping the troublesome question of the adequacy of the income data and broadening the definition of poverty to include other measures of deprivation than just income. As he puts it: ‘Restoring credibility to Australian poverty research involves demonstrating that those who are classified as poor are experiencing deprivation…and do not just report low income in a social survey.’[23]

I have no argument with this attempt to broaden the definition, for if we insist on using a concept like ‘poverty,’ it is probably best measured using a combination of indicators. Income is obviously important, but so too are signs of deprivation such as use of pawn shops to raise cash, or foregoing a child’s birthday party through lack of funds. What Saunders neglects to mention, however, is that research has already been done linking income to various measures of hardship, and the results are not encouraging from his point of view.

For a start, few low-income households say they suffer hardship, yet significant numbers of households on higher incomes claim to suffer substantial hardship. A 2001 report by the Commonwealth Department of Family & Community Services found that no more than one in six ‘poor’ households (with equivalised incomes below half the median income) reported any hardship, and fewer than one in eight reported ‘multiple hardship’. Half the households reporting multiple hardship had incomes above (often well above) the poverty level.[24]

There are two possible explanations for these paradoxical findings. One is that people are not reporting their incomes accurately, which is why income correlates so weakly with incidence of deprivation. The other is that experience of hardship has as much to do with how people behave as with how much cash they have coming in. Neither explanation helps Saunders much, for the first raises again the issue of the reliability of the income data (which he has done his best to ignore) while the second opens up the question of ‘behavioural poverty’ (which he does his best to deny). In his book he is scathing about CIS claims that people’s behaviour can push them into poverty,[25] but evidence from research on deprivation suggests he may have to rethink his position on this.

A further problem for him is that the number of households reporting serious hardship (things like being unable to afford to heat the home, or going without a meal through lack of money) is relatively small—around 3% of the population. These households are clearly ‘poor,’ but their numbers fall a long way short of the Smith Family’s poverty estimate of 13%, still less the ACOSS fantasy figure of 23%. The CIS has always maintained that, defined sensibly, the poverty rate in Australia is no higher than 5% and is probably lower than that, which means it is a manageable and targetable problem. Evidence from research on hardship and deprivation appears to bear this out.[26]

Clearly, Saunders’s book is not going to end the ‘poverty war’, despite the hopes expressed by his cheerleaders on the back cover. But this may be no bad thing. It was Karl Popper who taught us that criticism is the means by which science progresses. It is crucial that empirical claims should be rigorously challenged and that theoretical propositions should be fiercely tested, for only in this way can we progress beyond fallacies and falsehoods. Even if everything the CIS has said had subsequently been proved wrong, its critique would still have played a positive role in testing the propositions that Australia’s social science research community has been so unwilling to subject to critical scrutiny. As it is, much of the CIS critique has stood up remarkably well to onslaughts from opponents, and this should have strengthened our knowledge and understanding of the issues.

If Australia’s social science and social policy academics were not so wretchedly homogenous in their political leanings, some of the more exaggerated and less defensible claims about poverty put out by the welfare lobby and social policy researchers down the years would have been challenged long ago from within the Academy. As it was, the academics failed to do what they are paid to do and it fell to an independent think-tank to throw up the challenge. The fact that Saunders interprets this challenge as a declaration of ‘war’ speaks volumes about the complacency that runs through our universities, for in a genuinely critical and pluralistic academic environment, disputes like this would occur all the time. In a healthy academic environment, debate and argument is the normal business of intellectual life; only in the closed, smug atmosphere of the Australian academic cloister does intellectual challenge feel like warfare.

Out in the real world, of course, none of this matters very much. It always was an absurd conceit for Saunders to suggest that the fate of the poor depends on a bunch of academics proving the poor exist in large numbers and then getting the government to take their statistics seriously. In reality, the fate of the poor depends not on the outcome of a war of words over poverty, but on the dynamism of the private sector economy.

Since the mid-1990s, the living standards of rich and poor alike have been rising at quite an extraordinary rate in Australia. In just eight years, those on the highest incomes have seen their real spending power rise by 16%, and even at the bottom end of the distribution where virtually nobody is working, real incomes have risen by 12%.[27] Saunders and his allies argue there is little evidence that prosperity ‘trickles down’ as the wealth of our society grows,[28] but this is obviously nonsense given that the poorest households in the country have seen their living standards improve by one-eighth in eight years. Their growing affluence has not been the result of highly-funded research projects, of government consultation exercises, of the establishment of poverty targets or of teams of experts telling the PM what to do. Rather, it has been the result of a strong market economy doing what it does best, increasing everybody’s prosperity.