Surprises in poverty indicators

On May 15, 2008 the Centre for Poverty Analysis held its 34th Open Forum to discuss the recently released Poverty Indicator statistics from the Department of Census and Statistics.

Dr. Suranjana Vidyaratne, Director General of DCS presented the findings, while Ms Neranjana Gunetilleke a senior professional with CEPA responded.

CEPA welcomes the recent publication 'Poverty Indicators' findings from the latest Household Income and Expenditure Survey (HIES) 2006/07 by the Department of Census and Statistics (DCS).

The findings provide updated poverty figures for the country and enable comparatives of the previous calculations published in 2004, when the National Poverty Line for Sri Lanka was launched.

A poverty line provides a minimum rupee amount that a household requires to meet basic needs (food and non-food) every month. The setting of such a line provides information on the proportion of the population of a country/region that is identified as being poor.

The HIES collected data from a sample of households on income receipts and consumption of food and non-food items from July 2006 - June 2007. The survey was conducted in 19 districts across Sri Lanka, excluding 6 districts in the conflict zone where it was not possible to do so.

These figures are aggregated to district level taking account of the profile of households in each district.

As a measure of poverty, absolute consumption is a static tool giving a snapshot at a point in time and is therefore limited. It does not, for example, measure the movements in or out of poverty, or how vulnerable people are to shocks such as food price increases or natural disasters.

Poverty is a multi-dimensional problem which can be experienced on other levels beyond consumption, for example lacking access to employment, education or basic amenities such as electricity and safe water.

Results

Criticisms can be directed at the Poverty Line methodology but accepting and working within these limitations, the indicators provide a varied picture of poverty across Sri Lanka.

The trend shows an overall reduction nationally, across the urban and rural sectors but not within the estates; and across all districts for which data is available, except Nuwara Eliya.

This trend is no surprise and follows the steady downwards pattern since the 1990s, with the 1995/6 increase considered as something of a 'blip' - explained by the prevalence of drought and agricultural decline at the time.

Looking beyond these figures to the disaggregated district level data the picture is more complex. The only dramatic increase is in Nuwara Eliya, the only district with increases in the proportion of the population below the poverty line, increasing to 33.8%, twice the national level.

The most dramatic reductions were in districts which had experienced stagnation since the 1995/6 blip, such as Kurunagala (25.4% in 2002 to 15.4%) and Hambantota (32.2% in 2002 to 12.7%); and those districts which had slow recovery from 1995, most notably Puttalam (31.3% in 2002 -13.1%) and Matale (41.9% in 1995/6 to 18.9% in 2006/7). Most other provinces displayed fairly continuous drops in poverty.

Another blip?

Overall the figures look reassuring. It is important to understand the trends, assessing the reasons for success is as essential as those of failure. Following 1995/6 the data readjusted positively back to trend, and if 2006/7 is another blip the adjustment back will be negative. Preempting such a situation could be vital in influencing the policy direction.

In her response to the DCS presentation on the Poverty Indicators, Ms Gunetilleke of CEPA identified several key areas of importance. Firstly the urban poverty figures stand out as considerably lower than rural and urban sectors.

While they remain low relative to the estate and rural sectors at 6.7%, the decline is slow and primarily due to reductions in Colombo and Gampaha districts.

Absolute figures of the number of urban match figures of some of the poorest districts. Urban sector poverty is generally overlooked in the research community, but these results do warrant attention.

The estate sector continues to remain an area of concern, although the latest increase is driven by the Nuwara Eliya district figures. Much work in this field is directed at welfare improvements.

CEPA's study of poverty in the tea and rubber plantations in 2005 concluded that it was the structure of the industry that hampered living standard improvements and that these issues were as important as welfare ones. The figures for Nuwara Eliya remain a surprise. There is little to suggest why estate and agricultural livelihoods should have fallen in this district in comparison to others.

The explanation might be that the DS divisions of Haguranketha and Walapane which feature in the 100 poorest DS divisions are pulling Nuwara Eliya downwards; these areas constantly face natural disasters.

For those districts which had experienced slow recovery from 1995/6 and subsequently a marked improvement in the latest dataset; in Matale and Kegalle there is nothing to suggest a booming economic recovery that would be required to propel such drops in poverty.

The trends in Puttalam and Hambantota are also difficult to explain. One explanation for Hambantota could be the inflow of post tsunami assistance, which may also explain the low figures in Batticaloa and Ampara.

It is also possible that during the time the survey was conducted in these districts, the tsunami affected areas were experiencing an influx of migrant workers working on post tsunami reconstruction projects and whose consumption was subsequently captured within the household survey.

However, since the tsunami did not affect the whole district these explanations alone cannot account for the improvements revealed by the data.

Another possible explanation for the changes taking place in Hambantota, proposed during the discussion, were the populist activities of two opposing politicians in the district producing a cumulative effect on poverty figures.

There are a range of factors which will contribute to the trends in poverty in Sri Lanka. Issues such as migration and remittances may play a vital role in understanding overall poverty figures. World Bank statistics estimate that in 2004 $1.3 billion was sent home to Sri Lanka as remittances, approximately 7% of GDP.

They are therefore a considerable source of income for consumption purposes, and having a fuller understanding of these issues regionally would be helpful.

Caution has to be exercised when interpreting aggregated results from surveys. Overall these poverty indicators show the continued decline of poverty across Sri Lanka, but with disparities remaining across sectors and regions. When assessing poverty in Sri Lanka, these figures should not be viewed in isolation, but rather need to be understood through more micro level analysis.

(This Op-Ed reflects the views of the Centre for Poverty Analysis following an open discussion, and does not reflect the views of any one individual present at the Open Forum. The Centre for Poverty Analysis holds regular Open Forums on Poverty. For more information on our events and publications, please see www.cepa.lk or email us at [email protected])