WHICHEVER way you parse the data we have it shows that poverty headcount in Pakistan over the last decade and a half to two decades has decreased substantially. Initially, it was thought the data was not good enough, that it had been manipulated and so on, but even after multiple rounds of national surveys, the same trends are evident. And though the actual percentage of the poor may vary with the method one uses, the trend of falling poverty remains invariant. There must be something to this trend.

Poverty headcount, by the old line, has reduced to below 10pc in recent surveys. It is common practice, when poverty headcount goes below 10pc odd, to rebase the poverty line so that it gives some meaningful numbers. Social policy, if it has to work with less than 10pc odd of the population, is not as effective and/or useful. When all sorts of analyses confirmed that Pakistan’s poverty headcount had indeed gone below 10pc, the ministries of finance and planning, with help from the World Bank, decided to rebase the poverty line. This rebasing was announced a few weeks back. According to the new poverty line and numbers, poverty headcount is around 29pc of the population.

At the same time, the perception in the country is that poverty, if it has not gone up, has not decreased. How does one square this circle? There are other puzzles here too. While the poverty reduction trend seems to be robust, malnutrition and stunting incidence, especially in children, seem to be on the increase.

Why has malnutrition increased even as poverty has come down?

If the population is able to meet their basic caloric needs, as well as purchase other necessities, why are malnutrition and stunting incidence increasing? Are people choosing to eat and feed their children poorly? Why would that be the case? There are some systematic changes in buying patterns in terms of a shift from non-processed food to processed food, an increase in meat consumption compared to lentil consumption, but these do not explain the malnutrition increase phenomenon. This is a very important puzzle to resolve.

Infant mortality and maternal mortality numbers have also been, more or less, stagnating over the same period. If poverty has come down, why is it not translating into better health and longevity outcomes for people?

One possible explanation here is that health outcomes are not only tied to the income level of a household but to availability of good quality public goods: water and sanitation facilities, healthcare facilities, and environmental conditions. Even if the income of a household increases, they might still be drinking poor quality water or using pits for waste water disposal and/or living in an environment where solid waste is not collected from the streets.

We know that a lot of Pakistani children suffer from diarrhoea and have worms in their digestive tracts and one major reason for both of the above is the fecal-to-oral route. We also know that drinking water quality, across the country, has been deteriorating. So, stagnation in health outcomes might have to do with lack of provision of needed public goods. And it might not be possible, now, to move on infant and maternal mortality and health outcome issues without major investments in public goods provision.

An even more interesting issue is that we do not really know why poverty has come down in Pakistan. What have been the determinants of reducing poverty and what has been driving it? It is definitely not tied to GDP growth in Pakistan. Over the last 15 odd years only two to three were reasonable-to-high GDP growth years (2004-2007). In other years, growth rates have been quite poor. But poverty, even over slow-growth years, has continued to decline. At the same time, we have also seen increases in inequality in Pakistan. And the government has not been very active on the redistributive side as well. So, if the economy is not growing fast, and there is no redistribution of existing resources happening, how is poverty coming down?

There are a couple of promising hypotheses here. Some researchers think remittance flows have been increasing substantially and they might explain the reduction in poverty. This, to me, does not sound too promising an explanation. Remittance numbers are not that large, but more importantly, remittance flows are unevenly distributed across Pakistan and it should be possible, through careful analysis, to see if higher poverty reduction has been achieved in areas where remittance flows have been larger.

Some researchers think that it is growth of the informal economy, over the last decade and a half, that explains the reduction in poverty. Our GDP series does not capture the informal economy very well.

So, if there has been growth in the informal economy, it is possible to see reductions in poverty without seeing a significant connection with GDP growth rates. We need much more detailed micro level work here to see if growth in the informal sector is indeed what is driving the reduction in poverty.

Poverty has reduced but we do not understand why and we do not understand the movement, or lack thereof, in correlates. Why has inequality increased? How come poverty reduction and GDP growth rates are not related? Why has malnutrition increased even as poverty has come down? Why are we not seeing reductions in infant and maternal mortality and why are health outcomes not improving?

What other investments, in public goods, are needed to move correlates in a desirable direction? It is time the poverty debate in Pakistan moves beyond the numbers issue. We need to understand the dynamics of poverty and poverty reduction better. This is imperative for designing effective social-sector policies.

The writer is a senior research fellow at the Institute of Development and Economic Alternatives and an associate professor of economics at Lums, Lahore.

Published in Dawn, April 22nd, 2016