The poverty dominoes are poised for collapse once again. And as always, the core problem has to do with the definition, the calculation and the supporting statistics for poverty.

While releasing the Pakistan Economic Survey for 2013-14, Finance Minister Ishaq Dar announced that poverty was down to 12.4 per cent in 2010-11 as compared 17.2 per cent in 2007-08. According to the economic tsar of the country, the lower incidence of poverty stems from heavy investment in social safety programmes (for example, the Benazir Income Support Programme launched by his predecessors at his behest), better support prices for agricultural products, an increase in corporate and individual philanthropy, stronger inflow of remittances and the increasing number of women workers in the labour force.

But in the same breath, Dar went on to add that poverty should be computed on the basis of $2 per day -- instead of the $1.25 used previously -- which will show that 50 per cent of Pakistan’s population lives below the poverty line. And given the vast discrepancy between the two sets of figures, all his incredulous audience could do was nod disbelievingly as Dar segued into his usual improve-growth-reduce-poverty mantra.

The confusion intensified, once people studied the survey. According to the top notes of the World Bank’s Poverty Head Count Analysis 2014 printed in the Economic Survey, using the $1.25 level with the population figures of 2008, will render 21.04 per cent people poor, not the 17.2 per cent cited by the government in the survey. And raising the bar to $2 -- the level for middle income countries around the world -- will render 60.19 per cent poor, even if sticking with the 2008 population figures.

And this contradictory range of figures -- between 12 per cent, 50 per cent and 60 per cent -- shows what’s really wrong with our understanding of the incidence of poverty and the policy attempts to alleviate it.

Given that there hasn’t been an official census conducted since 1998, most population figures are either reverse engineered or educated guesswork. Consequently, all calculations predicated on them are invariably skewed.

But the most important issue in poverty calculation is to recognize how the poverty machine was recalibrated in the last decade and a half. And for the government to admit that unless it effects a technical correction to this system, reliable numbers on poverty will always elude policymakers and social scientists.

In the late nineties, for example, the Planning Commission decided to change the minimum calorie intake from 2,550 calories per day, per adult to 2,150 calories. (The justification was that calorie intake should be weighted for population -- women and older people needless -- instead of applying the average consumed by males between the ages of 20 and 39.) Immediately, poverty dropped from 32.2 per cent to 28 per cent.

And then the floodgates of number fudging opened wider. In 2002, the women-and-elderly justification was rubbished and the Planning Commission upped the average calorie consumption level to 2,350 (which is still used) and poverty came up to 30.6 per cent.

But when the PIHS figures for 2001-2002 came in, the Planning Commission increased its estimate regarding the poor for the same period to 32.1 per cent. (Other independent sources put the poverty number between 35.6 per cent and 38 per cent while the World Bank plumped for 37 per cent.) The reason for the lower government figures, analysts have pointed out since, was the fact that it had cut its sample size by five per cent and upped the poverty line. Significantly, the poverty line the Planning Commission used -- Rs 748 per capita per month -- implied a higher level of inflation that wasn’t borne out by even the Consumer Price Index numbers.

The massaging of the poverty figures continued well into the Pervez Musharraf years. The general had decided to prolong his stay indefinitely and he needed to deliver on the economic miracle he had promised in his first address to the nation. Thanks to the War on Terror dollars and a consumption-fuelled economic growth, promising signs were emerging. But to perpetuate the myth, Musharraf needed better poverty figures.

Obligingly, the statisticians delivered. To calculate the 2004-05 poverty estimates, they used CPI numbers instead of those thrown up by the survey of household spending or even the Sensitive Price Index. The reason the underlying figures become so controversial is because economists insist that the CPI (then calculated from prices for a specific basket of goods at various retail stores in 16 urban centres; now, from 40) doesn’t capture the consumption pattern of the poor. Further, they argue, given there is a higher incidence of poverty in the rural areas, the reliance on urban spending numbers is misplaced.

But the magic worked and Musharraf was able to serve up to the world and his detractors a very commendable 23.9 per cent. (Had Shaukat Aziz wavered from his consumption-led growth religion, he could have tightened the monetary policy. Not only would it have led to a drop in inflation, the number of poor would also have decreased further.) But, as it happened, the World Bank challenged the numbers in 2006 and insisted that inflation in Pakistan was far higher than it was stated to be and arrived at a poverty headcount of 29.2 per cent for 2004-05.

The steady decline in poverty in the years since -- from 22.3 per cent in 2005-06 to 17.2 per cent in 2007-08 and 12.4 per cent in 2010-11 -- would have been miraculous, if true. But truth is, even if the BISP and support prices argument were credible, the historical number fudging has cast a doubt over all poverty-related claims.

To his credit, Ishaq Dar has admitted the possibility of the numbers for 2007-08 and 2010-11 being off. He has now even set up a technical group to review the official methodology and to find out possible causes of variance in poverty numbers and recommend a final and conclusive set of figures.

However, even the corrected figures won’t mean much on their own unless similar time and attention is given also to the policy prescriptions Dar comes up with. His current insistence on economic growth and industrial growth as the panacea for what ails Pakistan shows a disturbing proclivity for the tried, tested and failed trickle-down policies that have rarely, if ever, made a dent in poverty anywhere. The finance minister would be well advised to come up with a more innovative approach to poverty alleviation. But can he?