JCTR - Key aspects in a poverty reduction strategy

KEY ASPECTS IN A POVERTY REDUCTION STRATEGY: CAN THE PRSP DEPART REMARKABLY FROM SAP?

As Zambia goes through the process of preparing its Poverty Reduction Strategy Paper, (PRSP), every one of the participants concerned with the process must, in the first place, have a clear idea of what each of the terms in the PRSP implies, namely, Poverty, Reduction, Strategy and Paper. Firstly, what is the concept of poverty? How should we define it? Secondly, in relation to poverty howsoever defined, what specifically shall we be reducing and for whom? Thirdly, how can we best achieve this reduction? In other words, what would be the appropriate strategy to use? And finally, how should this Paper on Poverty Reduction Strategy be prepared? It is not my purpose in this paper to dwell at length on all these issues since each of them would merit a paper on its own. I would only like to indicate some key considerations that should be taken on board in addressing these issues.

Poverty is the most visible characteristic of underdevelopment. It is the negation of development. It is the negation of human rights. It is the negation of the hopes and aspirations of human beings. It is the negation of opportunities for human beings to achieve self-realization. Since development is today regarded as a multi-dimensional concept, and since poverty is the negation of development, poverty too is multi-dimensional. It has many faces and facets. A recent World Bank document cites the following eloquent description of poverty given by a poor woman in Moldova:

“Poverty is pain; it feels like a disease. It attacks a person not only materially but morally. It eats away one’s dignity and drives one into total despair”. (World Bank, 2000).

An individual is poor if she falls below a certain poverty line. But the poverty line is not unique. Most often, poverty is measured by the percentage of the population falling below a money-metric poverty line. The World Bank for instance considers one dollar a day as the minimum income individuals should have so as not to be considered to be poor. In Zambia, the CSO has also developed similar poverty lines in kwacha terms. These poverty lines are based on the notion of a minimum income or expenditure that a person should enjoy. However, this is only an income or expenditure poverty line that can be used to gauge income or expenditure poverty. It represents one, but only one, facet of poverty. One can think of a host of other poverty lines that could show up other facets of poverty: knowledge poverty line, health poverty line, nutrition poverty line, shelter poverty line, water poverty line, voice poverty line, capability poverty line, and so on.

In Zambia, as per the latest available statistics, 73 percent of the population have incomes below the minimum level stipulated by the CSO: they are income-poor as measured by CSO’s income poverty line. Only 35 percent of the rural population have access to safe water; hence 65 percent of the rural population are water-poor. 59 percent of the children are stunted: they are nutrition-poor. 27 percent of the population have not had any schooling at all: they are knowledge-poor. And so on.

It must be noted that a person who is poor in one dimension of poverty need not be poor in another dimension. In particular, there is ample empirical evidence to show that income-poverty and non-income dimensions of poverty are not necessarily correlated in a strong way. As the eminent development economist Paul Streeten (1994) wrote:

“[A] unity of interests would exist if there were rigid links between economic production (as measured by income per head) and human development (reflected by human indicators such as life expectancy or literacy, or achievements such as self-respect, not easily measured). But these two sets of indicators are not very closely related.”

Hence in defining and measuring poverty, one has to consider a whole spectrum of poverty lines. Some of these poverty lines can be readily quantified as for example income poverty line. For some, good quantitative proxies can be used. For example, in defining poverty in terms of lack of access to safe water or health care, distance measures can be used. Some of the poverty lines are not easy to measure as in the case of voice poverty for example. But again some indirect quantitative indicators can be found.

It must be also noted that even poverty lines that admit of relatively easy quantification, need not be uniquely defined. As I have already stated, the income poverty line could be fixed at $1 a day, $2 a day, K30,000 a month and so on. It is based on the cost of a basket of minimum needs. This cost would depend on how the minimum needs basket is defined and what price indices are used.

Each income poverty line would yield its own estimate of poverty. For instance, on the basis of the CSO’s poverty line of K28, 979.4 per month, 69.2 percent of the Zambian population were estimated to be poor in 1996. If the poverty line of $1 a day were used, 72.6 percent of the population would have been estimated as being poor. And the estimate would have risen to 91.7 percent if the poverty line were raised to $2 a day. (World Bank, 2000). Further, a Study Fund Report by Miti et al (1999) stated that whereas the CSO estimate of poverty in Zambia in 1996 was 69.2 percent, this estimate dropped to 62.2 percent when the poverty lines were adjusted for regional price variations.

Again, any poverty line relating to access to some service or facility using a distance measure would depend upon what distance is considered to be convenient and not convenient. What for example is the minimum distance to be covered by an individual or household to access safe water beyond which that individual or household would be considered as lacking access? Is it more than 15 kilometers, more than 5 kilometers, more than 1 kilometer, more than 100 meters? Depending upon the distance that is used to define the poverty line, different estimates would be obtained for water poverty.

This difficult issue of defining poverty lines has to be tackled sagaciously. One must not only be pragmatic but also avoid political or other ulterior motives. If, for instance, a country wants to show that it has been improving because of the sound policies that it has put in place, it could adopt one set of poverty lines that would indicate lower levels of poverty. On the other hand, if it wants to exhort the donors to provide more funds, it could use another set that would show that the poverty situation is very desperate. This is one of the reasons why the issue of conceptualizing and defining poverty, among others, must be addressed through a wide participatory process that includes particularly the poor people’s own perceptions of poverty.

It must also be remembered that poverty reduction is a process. Hence changes in poverty levels need to be monitored and measured on the basis of consistent comparisons over time. For this we need to have a clear and sustainable definition of poverty – one that will not have to undergo changes too often or too much.

In Zambia, discussions on poverty have mostly revolved around the notion of poverty based on income or expenditure. That is how the Government itself earlier set the target of reducing poverty from 70 percent in 1996 to 50 percent in 2004. One can see the link between this notion of poverty expressed in terms of income deficiency and the Government’s constant emphasis on generating sustained income growth. The Government has kept stressing on the importance of generating growth in order to reduce poverty. But sustained income growth, while certainly being helpful, is not a guarantor of poverty reduction when poverty is defined more comprehensively to include non-income variables.

Even when poverty is defined narrowly in money-metric terms only, the relation between growth and poverty is not a straightforward one. On the basis of extensive empirical evidence, one can put forward at least three major propositions that, to my mind, are ideology-free and acceptable to all schools of thought. These are:

1. Higher the levels of sustained economic growth, greater are the chances of poverty reduction;

2. Higher the levels of prevailing poverty, greater is the difficulty of achieving sustained and high rates of economic growth;

3. Higher the levels of prevailing inequality, lower will be the impact of any economic growth on poverty reduction.

I will say more on these propositions when I discuss the kind of strategy that we chose for poverty reduction.

I now come to the next critical issue: what do we understand by poverty reduction? What should we seek to reduce? And for whom?

Two features of poverty are relevant to our discussion here. The first feature is that poverty does not manifest itself uniformly among countries, and within countries, among regions and socioeconomic groups. In some countries, income poverty is very high but poverty in terms of non-income facets is not commensurately high. In some other countries, income poverty is relatively low while poverty in non-income terms is much higher. Obviously, poverty reduction cannot mean the same thing for all countries. In the same vein, poverty may also manifest itself differently within a given country among different geographical areas and socioeconomic groups. For instance, Chipombo in the Central Province and Luangwa in Lusaka Province are equally poor. But child malnutrition is a far great problem in Luangwa than in Chibombo. On the other hand, access to a primary school is a serious problem in Chibombo while it is not a problem at all in Luangwa.[1]

In here, it is pertinent to note that non-income aspects of poverty are probably more important to address than income poverty. This is because income only has an instrumental value while many non-income variables also have intrinsic value. The value of income is that it is an instrument to access various material goods and services. But the value of education for instance is much more. Not only is it an instrument to achieve greater productivity and output, it also has the intrinsic merit of being a human right. If in fact, an individual’s right to adequate nutrition, health, knowledge, accommodation, potable water, sanitation and all other amenities that contribute to a decent living were guaranteed and as much as possible on the basis of her own choices, that individual would not need income at all!

In seeking to reduce poverty, therefore, one would need to gauge the overall level of deprivation that is suffered by the population and identify what specific areas contribute to this deprivation. This can be done by developing a composite index of poverty that includes not only income but non-income variables as well. The Human Poverty Index developed by the UNDP is one such index. In a recent study that I have done (Seshamani, 2000), I have formulated an Index of Deprivation based on 13 composite indices derived from 15 key poverty-related variables, both income and non-income. This Index has been calculated for all the 72 districts of Zambia. The value of this Index indicates the overall level of poverty and the values of the component indices point to the specific forms in which poverty most keenly manifests itself in any given district.

Since poverty has myriad ramifications, no Index can be comprehensive enough to capture all of them. The idea, however, is to develop an Index that includes a set of variables that is statistically sufficient[2] and robust[3].

The second feature of poverty is that poverty is not uniformly experienced by all individuals across geographical regions and socioeconomic groups. The experience can range from marginal to moderate to high to most severe. If for instance, a dollar a day is the income poverty line, an individual earning 95 cents is marginally poor while an individual earning 10 cents is highly poor and one with no income at all is most severely poor. Likewise, if 9 years of schooling defines the education/knowledge poverty line, one with 7 or 8 years of schooling is marginally knowledge-poor while one with no schooling is most severely knowledge-poor.

What would poverty reduction imply in such a situation? In other words, whose poverty should one reduce? If there were enough resources, one would eliminate everyone’s poverty. But when resources are limited, should one choose to lift those on the margin of the poverty line to bring them above it? Or, should one choose to raise those who are at the bottom of the heap to somewhere up but still below the poverty line? There is no easy answer to this question.

Often, a distinction is made between poverty reduction and poverty alleviation. When you lift people to bring them above the poverty line, it would be regarded as poverty reduction. When you lift people at the bottom some way up, it would be regarded as only poverty alleviation since those people continue to remain poor. But another way of looking at it is to think that in the former case you will have reduced the incidence of poverty while in the latter case, you will have reduced the severity of poverty. Thus both instances are only different versions of poverty reduction. So the question still remains: which version of poverty reduction should you adopt?

My own answer is contained in another paper that I wrote some time back (See Seshamani 200a). I will not repeat it wholesale here since many of you may have read that paper and those who haven’t can read it later. In a gist, I feel it is more useful to reduce the severity rather than the incidence of poverty since it would have the greatest benefits in terms of equity and empowerment.

The guiding philosophy here is that development has to be achieved by the people themselves, but they must be endowed with the power to achieve it. Hence, one must begin by empowering those who have no power at all to develop themselves and progressively move to those who are deficient in power in varying degrees. In other words, a bottom-up approach to poverty reduction would help to accelerate development in Zambia more than a top-down approach.

A concomitant approach that I have suggested is with regard to areas of spending by way of reducing severe poverty. This is the osmotic approach as distinguished from the trickle-down approach. In this approach, you need to identify a cluster of activities that have the greatest potential for reducing poverty and target resources thereon. While seeking to assist those in greatest need, you also seek to optimize the spread effects of spending on growth and development. Such activities include for instance assistance to women without support and their households, free primary education for girls, rural roads and micro credit for small-scale farmers. If resources are severely constrained, even focusing on or two of such activities can produce dramatic results in terms of reducing poverty and enhancing growth prospects. As more resources become available, one can identify second and third tier activities and so on. The osmotic approach requires using resources to reduce the depth and severity of poverty and to reduce inequity before going on to programs that are more likely to be distributionally neutral in their impacts.

In devising an appropriate poverty reduction strategy, one has to begin by taking stock of the prevailing situation. During the past decade, there has been no growth, poverty has increased in both monetary and non-monetary terms and inequality has continued to remain high.

The only major achievement the country has had was in containing the very high inflationary situation in the early nineties. Right now, stabilization is still a source of worry. Interest rates are high and the kwacha is on a slippery slope both domestically and externally.

In the early years of reform, a lot of time was spent in doing what critics of the erstwhile regime had urged Zambia to do, namely, diagnosing the crisis right (unsustainable fiscal deficits, huge losses of parastatals, etc) and getting the prices right (through removal of subsidies and controls). But Zambia also had to spend valuable time in doing one more thing - setting the relations right.

Relations with big brother Paul had been strained because for some time you had not been paying the interest on what you had borrowed from him. And later when you wanted to mend your fences with him, you did not have the money to pay him back. You could not borrow from Peter, John and Matthew because, since Paul was upset, Peter, John and Matthew also would not lend. So you had to say sorry to Paul and go down on your knees to plead with him to help you out of the situation. Paul then came up with an ingenious strategy. He would facilitate your borrowing from Peter, John and Matthew in order to clear his arrears so that you could start borrowing again from Paul who would then give the green signal to Peter, John and Matthew so that they could also start lending more again. But Paul agreed to do this only if you performed several Herculean tasks, otherwise known as fulfilling benchmarks, to his satisfaction. This program of just setting the relations right, also known as the Rights Accumulation Program, took three years and was successfully completed by the end of 1995.

I am not saying that these things were not necessary to do but in the process of doing all this, the rapidly deteriorating social dimensions in the country got largely ignored.

Having taken stock of the situation, we must now see what needs to be done. Three things are equally important to generate growth and reduce poverty: Adequate resources, sound policies and strong institutions. If resources are inadequate, not much can be done. Even if resources are there but the policies are misplaced, the resources can get frittered away with little development impact. But even if there are adequate resources and the policies are sound, poverty reduction is not likely to be achieved if the institutions are weak and ineffective. Weak institutions, in other words, can considerably undermine the impact of resources and policies.

In Zambia now, the least of the problems is the policy framework. But the biggest problem perhaps is not the paucity of resources but the weakness of institutions due cognizance to which has not been given. The weakness of institutions – and Government is one of the key institutions – is reflected in the country’s weak performance on governance indicators.

World Bank researchers Kaufmann, Kraay and Zoido-Lobaton (1999) have identified six major components of governance:

1. Voice and accountability which includes civil liberties and freedom of the press;

3. Government effectiveness which includes the quality of policy-making and public service delivery;

5. Rule of law which includes protection of rights and independence of judiciary; and

Kaufman, Kraay and Zoido-Lobaton have developed measures for the six components of governance and calculated them for 170 countries including Zambia. The data show that except in the case of the regulatory framework, Zambia’s position is relatively low among the 170 countries in all the remaining areas, especially control of corruption.

This is most likely the reason why in spite of what is a seemingly attractive policy framework in place for investment, major inflows especially of foreign investment in productive sectors and infrastructure development have not been forthcoming. According to the World Bank’s latest World Development Report 2000/2001, as of March 2000, Zambia had a score of 15.1 on Institutional Investor Credit Rating[4] and 58.8 on the Composite ICRG Risk Rating[5].

In devising a strategy for poverty reduction, therefore, serious attention will have to be paid to see how the various economic, legal and political institutions can be strengthened. The institutional framework is as important as the resource profile and the economic policy framework.

Let us consider one example to illustrate this. Assume that you have a fair amount of resources and the right policy of directing a good proportion of those resources toward health and education. But corruption could cause the resources to be diverted to the purchase of posh cars for management or to the election campaigns of politicians instead of to the purchase of drugs for patients and desks for schools. Even if this did not happen and drugs and desks were in fact purchased, corruption at this level may take the form of right tender procedures not being followed. As a result, the drugs and desks purchased may be of poor quality or of lower quantity than what would have been possible. Even if the drugs and desks were purchased in the right quantity and of the right quality, corruption at this stage could still prevent the drugs and desks from reaching the intended beneficiaries. The ultimate effect would be that there would be no visible development output corresponding to the allocated resources. And this would have happened because there was no proper institutional set-up to guarantee transparency and accountability in the use of resources and prevent resource leakage and fungibility.

Now I come to the relationship between growth, poverty and inequality. The three propositions spelt out in the first section suggest clearly that there is no recursive transmission of growth-generating processes in the direction of poverty and inequality reduction. It is perhaps a lack of due acknowledgement of this fact that has led to the conspicuous focus hitherto on attempts to promote growth without paying enough attention to equity issues and empowerment programs that could have enhanced the country’s chances of achieving growth. The result has been a failure on all fronts – growth, poverty and equity.

One of the major tasks of the PRSP should, therefore, be to see how a right balance could be achieved between all of them so that the country can generate poverty-reducing and equitable growth on a sustained basis. While maintaining a policy framework that would stimulate savings and investment needed for growth, resource allocation patterns should also allow for adequate interventions that would provide for social safety nets, effective social service delivery and human capital formation. The country has already reached such undesirable levels of poverty and inequality that even if a substantial growth is achieved but worsens poverty and inequality, such growth would not be desirable. It would even be preferable, in my view, at this stage to continue to have no growth at all if the only other alternative was to have any growth that further increases the number of have-nots and/or widens further the gap between the have and the have-nots.

But such need not be the case. One could simultaneously have growth as well as reduction in poverty and inequality. Indeed, wide empirical evidence shows that growth in general does have a positive net impact on poverty.

What then brings about pro-poor and inequality reducing growth? According to the World Bank, growth is the “outcome of countries’ initial conditions, their institutions, their policy choices, the external shocks they receive and no small measure of good luck.” Further, “Growth depends on education and life expectancy, particularly at lower incomes. For example it has been shown that female literacy and girls’ education are good for overall economic growth.” (World Bank, 2000, p. 49).

In order to generate pro-poor growth, one must focus on those sectors where the poor are concentrated. In Zambia, for example, if growth is generated through the development of trade and finance sectors while agriculture is neglected, such growth is not likely to reduce poverty much.

Zambia’s attempts to stimulate growth so far have been based on a policy framework in line with the Structural Adjustment Program or SAP. Since no growth has been stimulated in a decade, should the country then adopt a macroeconomic framework that radically departs from SAP? I do not think so. The SAP policy framework is based on empirically supported sound macroeconomic principles that include, among others, controlled growth of money supply, fiscal discipline, market-determined pricing and a liberalized environment for the private sector. However, by virtue of hindsight, what is needed is to avoid a dogmatic adherence to these principles and adopt a more flexible and pragmatic approach in implementing them.

The right strategy is one that is able to make a proper reckoning of tradeoffs. Stabilization and liberalization are desirable but are not ends in themselves; they are preconditions for growth. Sustained growth is desirable but is also not an end in itself; it is a precondition for poverty reduction and development.

In this regard, Stiglitz, who was until recently Vice president of the World Bank, has made several counter propositions against the orthodox manner of SAP policy implementation which are very relevant in the Zambian context. Some of these are:

· The focus on inflation has led to macroeconomic policies which may not be the most conducive for long-term economic growth;

· Budget deficits can be OK given the high returns to investment in such crucial areas as primary education and physical infrastructure, especially roads and energy.

· Macroeconomic stability is the wrong target. Minimizing or avoiding major economic contractions should be one of the most important goals of policy.

· Markets are not automatically better. Left to itself, the market will tend to under-provide human capital. Government must intervene.

· The dogma of liberalization has become an end in itself and not a means to a better financial system. Financial markets do not do a good job of selecting the most productive recipients of funds or of monitoring the use of funds.

The above propositions merit legitimate consideration in the Zambian case. The excessive monetary tightness has kept the cost of loanable funds high. The cash budget has constrained resource allocation to the social sectors. Credit has become out of the reach of small-scale farmers and businesspersons. Liberalization especially in the area of trade has caused firm closures. Clearly, while the policies are right, they need to be carefully retrofitted to ensure that they serve their purpose.

One further lesson that one can learn from past experience is that when a policy regime is changed, it can influence the parameters of private behavioral functions, i.e., the sensitivity of private agents to government decisions, or the functional form of private agent’s behavior. Economists call this the Lucas critique.[6] Clearly, the disastrous consequences experienced in the immediate wake of the liberalization of the agricultural markets in Zambia presents a validation of the Lucas critique. The Lucas critique also suggests the possibility of reciprocal interactions between the behavior of private agents and the government.

The above are some of the major factors to be considered in designing a poverty reduction strategy.

In this last section, I make some observations on the process of preparing the PRSP. I begin by citing two quotations from a recent IMF document:

“The new approach to poverty reduction…. builds on the traditional emphasis on macroeconomic structural soundness by also stressing the importance of country ownership of the poverty reduction strategy. The growing concern for country ownership, including through the involvement of civil society, is intended to reduce the risk of slippages in implementation as the countries themselves take greater responsibility.”

“To foster ownership, the PRSP will be drawn up by the government after broad-based consultations with stakeholders, including representatives of civil society and development partners, and with assistance with World Bank and IMF staff.”

Both the above quotations are taken from the same page (114) of the IMF document (IMF, 2000). The statements stress the need for country ownership of the PRSP. Ownership of any document requires effective participation in its preparation. But what seems to be suggested is only consultation of civil society by the government. There is a fundamental distinction between a consultative process and a participatory process. In the former process, there is only sharing of information by the government with civil society (and often even this information may be limited) but not a sharing of responsibility and decision-making. In a participatory process, government and civil society as also the intended beneficiaries would be jointly involved in deciding upon the preparation of the document so that they would be collectively responsible for its implementation.

Ownership of the PRSP by the country requires that it be prepared through a participatory process. A consultative process can only ensure ownership by the government and not necessarily by the country. It is therefore important to be clear whether the preparation of the PRSP is to be based on a participatory or a consultative process.

Further, what would happen even if the PRSP turned out to be a truly home-grown document prepared through a genuine participatory process involving the country’s think-tanks but is not finally acceptable to the IMF and the World Bank? Would the latter simply reject the document and refuse to fund programs under its umbrella? Or would they be prepared to sort out differences through an open dialogue?

To look at this issue from another perspective, would the IMF and the World Bank prefer to finance a PRSP that meets their expectations but which may not represent country ownership rather than finance a PRSP that has country ownership but does not meet their expectations? If such would be the case, the question would then arise as to why the country should undertake the PRSP process at all. It would be more efficient for the Government to simply sit with officials from the IMF and the World Bank and prepare the PRSP without spending time and money to involve the civil society in the process. This is what used to happen in the case of the Economic and Financial Policy Framework Papers or PFPs under SAP.

The key question then is: will the process of preparing the PRSP turn out in essence to be different from the process of preparing the PFPs?

Acocella, N. (1998): The Foundations of Economic Policy, Values and Techniques, Cambridge university Press, Cambridge.

International Monetary Fund, IMF (2000): World Economic Outlook, May 2000, IMF, Washington D. C.

Kaufmann, D., A. Kraay & P. Zoido-Lobaton (1999): Aggregating Governance Indicators, Policy Research Working Paper No. 2195, World Bank, Washington, D. C.

Lucas, R. E. (1976): Econometric Policy Evaluation: A Critique in K. Brunner and A. Meltzer (Eds): The Phillips Curve and Labor Markets, Carnegie-Rochester Conference Series on Public Policy, North Holland, Amsterdam.

Miti et al (1999): A Comparative Poverty Profile: Adjusted vs Unadjusted Expenditure, The Study Fund, Lusaka.

Seshamani, V. (2000): Some Pertinent Considerations in the Preparation of Poverty Reduction Strategy Paper for Zambia, Jesuit Center for Theological Reflection, Lusaka.

------------------ (2000a): Deprivation in Zambia: District level Rankings based on an Index of Deprivation, UNICEF, Lusaka.

Streeten, P. (1994): Human Development: Means and Ends, American Economic Review, Vol 84, pp. 232-237.

World Bank (2000): World Development Report 2000/2001, Oxford University Press, Oxford.

-------------- (2000a): Voices of the Poor: Can Anyone hear Us? Oxford university Press, Oxford.

[1] This is based on my recent study of deprivation in Zambia’s districts

[2] A statistically sufficient index is one that conveys all the essential information relating to the phenomenon that the Index seeks to represent.

[3] The term ‘robust’ is used here in the sense that the values of any one component will not unduly influence the value of the aggregate Index.

[4] The Institutional Investor Credit Rating ranks, from 0 to 100, the probability of a country’s default. A larger number indicates a lower probability of default.

[5] The Composite ICRG Risk Rating is based on 22 components of risk. It also ranges from 0 to 100. A rating below 50 indicates very high risk while a rating above 80 indicates a very low risk.

[6] It was advanced by R. E. Lucas in a paper written in 1976.