The question of poverty reduction in Pakistan
Official claims of a major breakthrough in poverty reduction in Pakistan do not stand up well to close scrutiny, argues Akmal Hussain
Umair Mohsin / Flickr
Poverty in Pakistan is rooted in an acutely unequal distribution of productive assets and an associated asymmetric structure of power. This power structure distorts markets and state institutions in such a way that they systematically discriminate against the poor with respect to access over resources, public services and governance decisions which affect their immediate existence.1
Poverty occurs when individuals in a fragmented community are locked into a nexus of power which deprives the poor of their actual and potential income. For example, in rural Pakistan it has been estimated that poor peasants are losing as much as one-third of their income due to asymmetric markets and local institutions of governance.2 At the same time, as many as 57.4 percent of the extremely poor peasants who have taken loans from the landlord work on his farm without any wages at all, and 14 percent work at a wage that is less than half the market wage rate.3
Our recent research shows that as much as 65 percent of the poor population in Pakistan is suffering from ill health, which is an important factor in the systemic process of poverty creation.4
Given the structural nature of the poverty phenomenon in Pakistan, a mere acceleration in GDP growth cannot, in itself, be expected to significantly reduce poverty quickly. For substantial poverty reduction to occur the structure of power, the asymmetric markets, and the distribution of productive assets have to be reconfigured in favour of the poor.
At the same time, the composition of GDP growth, which is determined by these structural factors, would have to be restructured in favour of the poor. On the contrary, the fact is that the government, during the last six years, has not even begun to address the structural basis of systemic poverty in Pakistan. Furthermore, the evidence shows that GDP growth during the last six years has been pro-rich rather than pro-poor.
It is remarkable, therefore, that the government has come up with a claim that it has overcome one-third of Pakistan's poverty problem within just four years. It may be useful to subject this claim to scrutiny in this brief note.
The government has claimed that poverty has declined from 34.46 percent of the population in the year 2001 to 23.90 percent in 2005, i.e. a ten percentage point reduction in poverty, which means that one-third of Pakistan's poverty problem had been overcome within a period of four years. If this were true, then the government has brought off one of the most dramatic feats of poverty reduction, within the shortest period of time, in the history of the developing countries over the last 100 years.
A number of questions, however, arise with respect to the government's claim. It is important for independent experts in civil society to determine whether or not poverty has declined substantially during the tenure of this government. If poverty has not declined substantially, then a redesigning of the government's economic strategy would have to be undertaken in order to achieve a high GDP growth that is both pro-poor and sustainable.
Questioning the government's claim of poverty reduction
The government's claim can be questioned at a statistical level in terms of three aspects of poverty reduction:
Sources and pattern of growth: The government's claim of having wiped out one-third of the poverty problem in Pakistan, over a period of four years, is not consistent with the sources of GDP growth and the pattern of agriculture growth.
For example, the agriculture sector, where the predominant proportion of Pakistan's poor population subsists, has shown sharp fluctuations in agricultural output which would be expected to increase, rather than decrease, poverty. During the period 1999-2000 to 2004-05, the index of production for major crops was substantially less than in the base year in two out of the six years. For food crops, in four out of the six years, the index of production was substantially lower than in the base year (1999-2000).5
When there is a bad harvest in any particular year it throws marginal farmers into poverty. They are forced to borrow money to buy food, and are unable to invest in seed, water and fertilizer for re-constituting the production cycle the following year. Thus, such a high frequency of bad harvests in the major crops sector means that marginal farmers have fallen into poverty. This suggests an increase, rather than a decline, in poverty. Moreover, the sharply increased inflation rate of wheat and other food items means that the poor have fallen deeper into poverty. (While the average inflation rate in 2004-05 was 9.3 percent the food inflation rate was 12.5 percent).6
The growth in the large-scale manufacturing sector in Pakistan has been predicated on the output growth of just four industries: automobiles, consumer electronics, textiles, and construction. Three out of these four industries neither produce goods for the poor, nor employment. This is because they are highly capital intensive, and whatever little employment they generate is for the highly skilled. Moreover, even though there was some increase in installed capacity, most of the growth in the large-scale manufacturing sector is attributable to better utilization of existing production capacity.7
The third component of GDP growth is the services sector whose growth has been predicated primarily on banking (15.2 percent growth in the year 2003 and 14.4 percent growth in 2004) and telecommunications (68.8 percent growth in 2005). The growth of the commercial banking sector, which was a key determinant of overall GDP growth, cannot be said to be generating employment, income or credit for the poor. Similarly, telecommunications which, while providing mobile telephones to a much larger proportion of the middle and lower-middle class, provided neither substantial employment nor cheaper consumer goods to the poor.
Bias in poverty reduction estimate due to base year selection: Poverty level in the year 2006, which was a good harvest year, has been compared with the base year of 2001 which was a particularly bad harvest year. For example, in the year 2001 rice production declined by 19.2 percent compared to the preceding year, while in 2006 rice production increased by 10.4 percent compared to the preceding year.
Similarly, the growth rate of wheat was negative in the year 2001, while it was positive in the year 2006. Clearly the selection of 2001 as the base year would tend to give an upward bias to poverty reduction. The government and multi-lateral agencies, in the interests of intellectual integrity, ought to have done a sensitivity test to examine the magnitude of the bias due to the base year selection. We really have no way of knowing if, for example, the year 1999 was selected as the base year rather than 2001, whether poverty would have declined or increased in the year 2006.
Inflation rates and bias in poverty estimates: The magnitude of poverty reduction or, indeed, an increase in the poverty estimate is highly sensitive to the inflation rate applied in adjusting the poverty line. The poverty line used by the government estimate is Rs 723.40 per month in the year 2001-02, and Rs 878.64 per month in the year 2005-06.
Clearly, a higher inflation rate estimate would mean a higher poverty line and, therefore, a higher level of poverty in the year 2005-06. The government has not indicated what would be the magnitude of poverty reduction if a higher inflation rate, calculated by independent economists, had been applied. Moreover, the official inflation rate estimate applied to the poverty line is an average inflation rate for the population as a whole. If the inflation rate specific to the basket of goods consumed by the poor (in which food expenditure has more weight) had been estimated it would have been much higher and, therefore, the poverty level in the year 2005-06 would have been much higher as well. Moreover, the costs paid by the poor on services such as health, transport and credit for purchasing food during bad harvests is not even included in the government's inflation rate estimate.
Poverty reduction estimates and definition of poverty
The government's claim to have wiped out one-third of the poverty problem in Pakistan over a four-year period is based on a starvation level poverty line of Rs.878.64 per month. The human functioning of an individual, as Professor AK Sen has argued, depends not just upon food but also upon access to a range of services such as education, health, clean drinking water, sanitation, accommodation and transport. If these services were included in the definition of poverty, then not only would the magnitude of poverty be much higher, but, in view of the poor performance of the government in the provision of these services, poverty estimates would have registered an increase rather than a decline.
Poverty and empowerment: As my work for the NHDR has shown, the poor are losing one-third of their income due to asymmetric access to markets, for outputs and inputs, which are distorted by the local power structure. Apart from this, the vulnerability of the poor to crimes committed against them by individuals and government officials, as well as lack of access to justice, are major sources of income loss for the poor.8 These factors are, in many cases, a reason for the poor being locked into bondage to the local power holders, whether they be large landlords, government officials, politicians or mafia bosses. Genuine poverty reduction would mean liberating the poor from this nexus of power which perpetuates their poverty. This change in the power structure, so vital for poverty reduction, was not on the government's policy agenda during the period in which dramatic poverty reduction is being claimed.
1 See: Akmal Hussain, A Policy for Pro-Poor Growth, published in: Towards Pro-Poor Growth Policies in Pakistan, UNDP and PIDE Symposium proceedings, published by The UN System in Pakistan, 2003, Islamabad. Page 61.
2. Akmal Hussain, with inputs from A.R. Kemal, Agha Imran Hamid, Khawar Mumtaz, Ayub Qutub: Poverty, Growth and Governance, UNDP, Pakistan National Human Development Report 2003, Oxford University Press, 2003, Karachi. Page 78.
3. Ibid. Table 14, page 63.
4. Ibid. Pages 69 to 71.
5. Pakistan Economic Survey 2005-06, Economic Advisor's Wing, Finance Division, Government of Pakistan, 2006, Islamabad. Statsitical Appendix Page 21.
6. Ibid. Page 119.
7. Annual Report 2004-05, Volume-I, Review of the Economy, State Bank of Pakistan, 2005, Karachi. Page 43.
8. Akmal Hussain, with inputs from A.R. Kemal, Agha Imran Hamid, Khawar Mumtaz, Ayub Qutub: Poverty, Growth and Governance, UNDP, Pakistan National Human Development Report 2003, op.cit. Pages 62 to 69.
Akmal Hussain is an eminent Pakistani economist.