SOME PERTINENT CONSIDERATIONS IN THE FORMULATION OF A POVERTY REDUCTION STRATEGY PAPER (PRSP) FOR ZAMBIA

It is common knowledge that poverty is a pandemic phenomenon in Zambia. Until recently, the figure of 70% used to be quoted as being the incidence of poverty. This figure, calculated on the basis of a Kwacha-based Poverty Datum Line determined by the Central Statistical Office, is supposed to have risen to nearly 73% in 1998 as per the Living Conditions Monitoring Survey conducted that year. And a national target had been announced at the highest official level to reduce this incidence to 50% by the year 2004.  Currently, the preparation of a Poverty Reduction Strategy Paper  (PRSP) is under way as per the requirements of the IMF’s Poverty Reduction and Growth Facility (PRGF) that has substituted its erstwhile Enhanced Structural Adjustment Facility (ESAF).

The purpose of this present paper is to suggest that it may be highly inadequate to focus attention on poverty reduction in the sense of just bringing down the number of persons living below a poverty datum line defined in kwacha terms. Indeed, in the Zambian situation it may not even be the most imperative goal to achieve. Any PRSP that aims primarily to reduce such a money-based incidence of poverty, even if it be on a sustained basis, could end up largely misdirecting itself in the larger context of enhancing the country’s capacity for achieving sustainable human development.

Achieving a reasonable target in respect of poverty reduction even in this sense may seem a difficult task but in fact it is not the most daunting challenge. One could think of far more daunting but also far more meaningful challenges.  Addressing these challenges would require setting several other targets than just lowering the percentage of the population that lives below a money-metric poverty line.

At the outset, let me suggest the areas where targets would need to be set for an effective poverty reduction program. I shall then provide a rationale for these target areas in terms of the theory and empirical work that exist in the literature. Finally, an attempt will be made to prescribe a set of quantitative targets that we must seek to achieve simultaneously if sustainable human development is to become a reality for Zambians.

The target areas for an effective poverty reduction program are:

·        Extreme poverty

·        Access deprivation

·        Inequality

·        Empowerment

The above areas are not mutually exclusive. What happens in one area will impact on the rest.

Extreme poverty

Poverty is not a homogeneous phenomenon. The poor are not all equally poor; some people are poorer than others. There are in fact layers of poverty ranging from marginal deprivation to the most acute forms of deprivation.  Two concepts that capture these poverty differentials are the intensity or severity of poverty and the depth of poverty.

The intensity of poverty refers to the extent to which the incomes of the poor lie below the poverty line. The depth of poverty looks at the degree of inequality among the poor. The intensity and depth of poverty are more difficult issues to address than the mere incidence of poverty. If the problem relating to a country’s poverty situation was merely one of incidence and not of severity and depth, it could perhaps be surmounted without too much difficulty.

Unfortunately, in the Zambian case, the poverty problem is not merely one of high incidence but also one of high severity and depth. According to the Living Conditions Monitoring Survey 1996, while the incidence of poverty was nearly 70%, the intensity of poverty was 34% and the depth of poverty was over 50%. It is these high levels of intensity and depth of poverty that militate against development.

Let me try to explain the above propositions in the context of the data from the Living Conditions in Zambia (1998) report. In this report, persons living in households with monthly Adult Equivalent Expenditure (AEE) of less than K32,861 were considered to be extremely poor. Those in households with monthly AEE of less than K47,187 but equal  to or greater than K32,861 were considered moderately poor. Persons in households with monthly AEE equal to or greater than K47,187 were considered non-poor.

The adult equivalent scales were worked out on the basis of a food basket that would ensure minimum calorie and protein requirements for persons in different age groups.

In 1998, 73% of the population lived in households with monthly AEE below K47,187. 15% lived in households with monthly AEE below K47,187 but not less than K 32,861. These were the moderately poor. 58% of the population lived in households with monthly AEE below K32,861. These were extremely poor.

The extent to which the AEE falls below K47,187 denotes the intensity of poverty. But the fact that, of all the households wherein the AEE falls below K47,187, 79.5% of them had incomes below K32,861 signifies the high depth of poverty.

Let us now suppose that a Poverty Reduction program enhances on a sustained basis the monthly AEE by K2000 in 100,000 households in Zambia. Let us consider three hypothetical ways in which this happens:

1.      The 100,000 households benefiting from the program already have monthly AEE between K45,200 and K47,000.

2.      The 100,000 households benefiting from the program already have monthly AEE between K30,900 and K32,000.

3.      The 100,000 households benefiting from the program already have monthly AEE between K25,000 and K28,000.

Assuming the population in 1998 was roughly 10 million and that 7.3 million of these were poor:

In the first case, the incidence of poverty would have been reduced by about 1.4%, moderate poverty would have come down by 1.4% but extreme poverty would have remained at 58%.

In the second case, the incidence of poverty would not have changed; moderate poverty would have increased by 1.4% to 16.4% and extreme poverty would have come down by 1.4% to 56.6%.

In the third case, the incidence of poverty as well as moderate and extreme poverty levels would have remained the same.

If your poverty reduction target was set only in terms of the incidence of poverty, you could claim some success only in the first case: poverty would have come down from 73% to 71.6%. But, it would have done nothing to change the intensity and depth of poverty. What is more, it would have worsened income distribution: those in extreme poverty would now constitute 81% instead of 79.5% of the poor. As we shall explain, the capacity of a country to achieve economic growth (which in turn could help reduce poverty) is very significantly and inversely related to prevailing inequality in distribution.

In the second case, the incidence of poverty is unaffected. But statistically, you could still lay claim to some success in that some of the extremely poor would have graduated to moderate poverty. The intensity of poverty would also have consequently been mitigated to some extent and there would be some improvement in the middle range of income distribution.

In the third case, the statistics relating to incidence, moderate and extreme poverty remain unaltered.   From a political perspective, this would not look attractive since there would not seem to be anything tangible to announce by way of achievement in statistical terms. But this case has other merits. It would have reduced overall inequality in distribution more than in the second case by improving the lower income ranges. It would also have the greatest impact on empowerment.

It is accepted today that people cannot be developed; they have to develop themselves. But in order to develop themselves they need to be empowered. Those who are not poor have enough empowerment. The moderately poor have some empowerment but not enough. The extremely poor are severely short on empowerment. If, therefore, the development capacity of a country is determined by the level of empowerment, then, following the law of diminishing marginal utility, for a given increase in income, the greatest addition to utility in terms of raising overall empowerment levels would be achieved if that income is spent on those at the lowest levels of extreme poverty than on those on the fringes of the Poverty Datum Line.

Thus, if the purpose of poverty reduction is to achieve concomitant enhancement in the  country’s development potential, the third situation is better than the first. In the first, case, the strategy seeks to compress poverty downwards from above, while in the third case, the strategy aims to compress poverty upwards from below.

In the first case, the impact on distribution is negative and the empowerment impact is the least. In the third case, there is the greatest positive impact on both distribution and empowerment.

Ideally, one should enable those in moderate poverty to graduate out of poverty and those in extreme poverty to graduate to moderate poverty and eventually out of poverty.  But given the resource constraints, this may not be possible to achieve simultaneously.

The appropriate strategy for an effective and equitable PRSP, then, should be to focus attention primarily on those who are most extremely poor and enable them to move progressively to less extreme levels of poverty. And as the resource situation improves, one can also address the less poor groups.

Access deprivation

Removing income deficiency in itself cannot relieve suffering and improve human well being. Income only represents potential command over material provisions that contribute to good living conditions. But if such material provisions are not readily available or accessible, then income does not have much value: it is little more than pieces of paper in one’s pocket. Poverty reduction must therefore mean reducing deprivation in access to basic material goods and services such as food, housing, primary health care and education, safe water and sanitation facilities, etc.

Access deprivation is a serious problem in Zambia especially in the rural areas. A significant number of households have to cover 16 kilometers or more to access potable water, a health clinic or a primary school. Many areas have virtually no toilet facilities or garbage collection facilities. Roads, public transport and access to markets also constitute major problems.

The PRSP must therefore specify targets to achieve in terms of reducing access deprivation.

Inequality

Any strategy that seeks to generate growth and reduce poverty without consciously seeking to reduce inequality will not achieve much success. A lot of empirical work points to a negative relationship between prevailing inequality and subsequent growth.  Recent research by economists, some of them from the World Bank (e.g. Alesina and Rodrik, 1994; Deininger and Squire, 1996) show that where initial inequality in respect of income, education, and assets, notably land holdings, is high, growth does not easily occur.

Indeed, inequality seems to affect growth even more than the political system. India and Philippines have been democracies with relatively high inequalities and have consequently not achieved very high growth rates. On the other hand, Korea and Taiwan, although not democratic, had lower levels of inequality and succeeded in achieving impressive rates of growth.

A crucial conclusion of empirical research is that a historically unequal situation might perpetuate itself unless changed by a pro-active government policy such as asset redistribution.

Inequality is very high in Zambia. The 1996 LCMS stated that the Gini Coefficient (a widely used measure of inequality) was 0.5. This is a rather high value of the ratio.

Another measure that indicates the degree of inequality is the Kuznets Ratio. This measure looks at the ratio of the share of the richest 20% of the population to the share of the poorest 60%. A value of this ratio in excess of 2 is deemed to indicate high inequality. According to the 1996 LCMS data, the richest 20% in Zambia had 67.8% of the income while the poorest 60% had only 16.2%. The Kuznets Ratio thus works out to a very high 4.2.

It is unlikely that high and sustained economic growth in Zambia can be easily promoted in the face of such inequality of incomes.

Empirical research also shows that a main reason why inequality tends to beget more inequality is the unequal access to credit. The poor cannot easily access credit owing to little or no wealth to provide as collateral and hence continue to languish in near or below subsistence state. The rich, on the other hand, have easy access to credit and hence are able to build up further on their already substantial wealth. Any poverty reduction program must therefore encompass provision of credit to the low-income groups.

But inequality exists not only with respect to incomes and credit, but also in respect of distribution of access to various other facilities. Disparities with regard to access exist on the basis of gender and geographical characteristics. Female-headed households in general have lower access to various facilities than male-headed ones. Inter-provincial and intra-provincial and rural-urban differences also exist. For instance, 1% of the households on the Copperbelt have no toilet facility whereas 73% of the households in Western Province have no toilet facility.  36% of the households in Lusaka Province have electricity but only 1% of the households in Eastern and Northwestern Provinces have electricity. And so on. Again, within Western Province, while the incidence of extreme poverty is 66% in Kalabo, it is 97% in Lukulu. Or in Northwestern Province, in Solwezi, the percentage of population with no education is 25 while it 54 in Mwinilunga. One could cite a number of similar examples.

But poverty is not just about percentages alone. The poor are human beings and every human life is valuable. So absolute numbers are important. A given area may have a high percentage of poverty but in terms of absolute numbers, the population involved may be relatively small. On the other hand, a smaller percentage could go hand in hand with larger absolute numbers. For instance, Table One looks at the incidence of poverty in Zambia’s provinces and compares it with the distribution of the poor among the provinces in 1998. 

Province	Incidence of poverty (%)	% distribution of the poor
Central	76.8	10
Copperbelt	65.0	18
Eastern	80.3	13
Luapula	80.9	7
Lusaka	52	15
Northern	81.1	12
Northwestern	75.8	5
Southern	75.8	13
Western	89.2	7

Table One

It is clear from the above table that the density of poverty within provinces (seen from the figures in the second column) is not necessarily correlated with the concentration of the poor among the provinces (seen from the figures in the third column). For instance, Western Province has the highest density of poor population (89.2%) but has the second smallest concentration of Zambia’s poor (7% which is next only to 5% in Northwestern Province).

If one paid special attention to the concentration of extreme poverty, then the contrast becomes even more stark. According to available data, Western and Luapula provinces contain only 5% of the extremely poor while Northwestern contains just 4%. On the other hand, the Copperbelt has 22% while Lusaka has 18%.

In terms of resource allocation in the poverty reduction programs designed under the PRSP, therefore, provinces such as the Copperbelt and Southern Provinces may require larger allocations as compared to Luapula, Western and Northwestern provinces.

Empowerment

Empowerment, be it at the individual, household or community level, denotes the creation of capacity to understand what can best contribute to one’s welfare, to make decisions accordingly and to implement those decisions.  It denotes, in other words, the ability of individuals, households and communities to exercise greater control over their life and resources.

Such control is an important precondition for development. But empowerment in this sense cannot exist unless an individual crosses a certain minimal threshold of physical and mental health, knowledge and access to material provisions.   This becomes the link to poverty reduction programmes.

In the Zambian context, empowerment requires access to land and credit, food, basic education and health, safe water and sanitation and social safety nets.  The PRSP therefore must pay due attention to these critically important elements.

Quantitative targets

In the light of the foregoing discussion, it must be clear that any meaningful poverty reduction programme cannot be related solely or even chiefly to a target for reduction of a money metric incidence of poverty. It needs other targets to guide its activities. We suggest in Table Two some of these relevant targets.  Obviously, the list is not exhaustive but it is merely indicative of the kind of targets that an effective PRSP must envisage for attainment within a specified time horizon.

Indicator	Medium term target	Longer term target
Extreme poverty	30% by year 2005	Eliminate by 2010
Clean water	70% households in rural areas by year 20005	100% Zambian households by year 2010
Distance to various amenities (schools, health centers, markets, etc)	Distance to no facility exceeds 15 kilometers by year 2005	All facilities available within 5 kilometer distance by year 2010 and within 1 kilometer distance by year 2015
% share of food in household expenditure	Not more than 40% for rural households by year 2005*	Not more than 25% for all Zambian households by year 2015*
Gini coefficient	0.4 by year 2005	0.3 by year 2010
Kuznets ratio	2.5 by year 2005	1.5 by year 2010 and 1.2 by year 2015
Per capita income	At least US$ 365 per annum by year 2005**	At least US$ 730 per annum by year 2015**

*   A percentage share in excess of 35 is deemed to signify high poverty.

** These figures are based on the World Bank’s criteria of $1 a day and $2 a day.

Table Two

Growth and Poverty Reduction

In Table Two above, the last proposed target on per capita income is linked to the achievement of respectable and sustained growth rates. This in turn is closely linked to the macroeconomic framework and policy instruments used. These are spelled out in the successive short term Economic and Financial Policy Framework Papers (PFPs) that are formulated by Zambia in cooperation with the IMF and the World Bank.

The PRSP is being prepared in the context of the IMF’s Poverty Reduction and Growth Facility (PRGF). The PRGF has supplanted the former Enhanced Structural Adjustment Facility (ESAF) but it has not substituted the PFP.  In other words, the IMF will presumably continue its own monitoring of Zambia’s policies through the PFP.

Will the PRSP then turn out to be just one big “Appendix to the PFP”? This question is reminiscent of the earlier criticisms that used to be made since the latter half of the 1980s of Structural Adjustment Programmes (SAPs); namely, that considerations of poverty reduction and social dimensions of adjustment at best appeared as add-ons to the SAPs inserted as only an after thought.

In fact, in one of my own papers written some ten years ago, I had shown how the few activities related to poverty reduction and development appeared at the tail end of the PFP with none of them bearing an asterisk. And an asterisk against an activity indicated priority!  We certainly must now be on guard against the danger of the PRSP also becoming a similar add-on to the PFP.

Indeed, if the PRSP is to serve as an effective guide to development, the PFP itself will need to be re-examined in terms of some of its content.  Implementation of the SAP in Zambia under the umbrella of the PFP so far has revealed some costly tradeoffs in terms of growth and development. 

For example, one such tradeoff is between stabilization and growth. Tight monetary and fiscal policies enabled Zambia to achieve impressive domestic stabilization but apparently such policies constrained the supply-side factors related to growth.

Another tradeoff is between the fiscal stance and development expenditures. The elimination of budget deficits and the achievement of modest domestic surpluses have caused inadequate spending in areas of priority human concerns such as primary education and primary health care. Such trends would obviously militate against the achievement of the kind of targets that we have set out in the Table One.

The right approach, therefore, would be not to prepare a PRSP that is subsequently appended to the existing PFP but to overhaul the PFP itself.  Then it could be consistently dovetailed it into the PRSP once the latter is prepared. In other words, Zambia’s programme of development would be guided in the future by the PRSP of which the PFP would be but an integral part.

Concluding note by JCTR

This paper is offered as a contribution by one civil society partner to the debate over the design of the Poverty Reduction Strategy Paper for Zambia.  We feel that a deeper analysis of what poverty reduction means is called for, one that moves beyond numeric targets and sectoral lists.  Professor Seshamani recommends taking a PRSP approach that:

·        looks at the intensity and depth of poverty and addresses programmes to deal with those parts of society most profoundly affected by poverty

·        reduces deprivation of access to basic necessities

·        addresses inequalities (e.g., income, gender, geographic) that hinder development

·        empowers people by creating capacities to understand and decide on future directions

Moreover, he suggests:

·        a series of targets, with indicators, that aim to implement these recommendations

·        a re-thinking of the relationship between growth and poverty that calls for a role for the PRSP that is more influential than the traditional PFP

We call for debate over these recommendations and suggestions, and invite comments aimed to improve the effectiveness of civil society’s role in the design, implementation and monitoring of the Zambia’s PRSP. 

Prepared for JCTR by
Professor Venkatesh Seshamani
Department of Economics
The University of Zambia
June 2000

To CSPR

Home | Information | Activities | Updates | JCTR Bulletin | Networking |  Cost of Living | Publications