Given the crisis of poverty in Pakistan and a number of other developing countries, for the first time in history economists, governments and multilateral donor institutions are united in attempting to achieve poverty reduction through economic growth. Most continue to believe in the conventional wisdom that GDP growth is sufficient to overcome the problem of poverty. By contrast, I have argued in these columns and in my published work over the last few years, that we must address not just GDP growth but the level and structure of economic growth that is consistent with rapid poverty reduction. Thus it is only when the pace of poverty reduction becomes a central concern that the content and structure of economic growth becomes a vital consideration in designing the growth strategy. It is only then that a deeper understanding of the relationship between growth and poverty reduction becomes necessary. (See for example my 1994 book titled: Poverty Alleviation in Pakistan and more recently my work embodied in the UNDP, National Human Development Report 2003). An important aspect of the pursuit of pro poor growth is to develop a quantitative method of measuring pro poor growth. This had eluded those of us who were engaged in designing and implementing strategies for poverty alleviation. Now in a seminal paper Nanak Kakwani and Hyun H. Son (January 2004) have made the break through in designing a simple but powerful method of measuring pro poor growth. In this article I will briefly outline some of the features of the Kakwani and Son measurement method, in the context of the concept and strategy of pro poor growth.

For over two centuries it was believed that economic growth is essentially a means of enhancing state power. Half a century ago, however, with the emergence of newly independent countries, attention focused on economic growth as a mechanism of shifting ‘developing’ countries out of poverty. It was thought that growth reduces poverty and therefore all that was required for overcoming it was to maximize the GDP growth rate. (This was the notion of poverty reduction as a “trickle down” effect of GDP growth). This indeed was the proposition that determined the design of the first three five years plan of Pakistan. Even today when the concept of pro poor growth is being propounded, the “trickle down” theory continues to hold sway. For example, Ravallion and Chen (2003), on the basis of their empirical work have defined pro poor growth simply as one, which reduces poverty. Yet the real issue is not the fact of poverty reduction (which GDP growth in most cases will achieve), but the magnitude and pace of poverty reduction in the growth process.

The importance of the Kakwani and Son (2004) contribution lies in the fact that in designing a method of measurement they have made the concept of pro poor growth more precise. They define pro poor growth as one in which a greater percentage of the increase in benefits goes to the poor relative to the non-poor. This definition is translated into a quantitative index, which Kakwani and Son call, “poverty equivalent growth rate” (PEGR). They demonstrate that the proportional reduction in poverty is a monotonically increasing function of the PEGR. i.e. The larger the PEGR, the greater the proportional reduction in poverty. With respect to economic policy the strategic objective is thus clarified: It is not just maximization of the GDP growth but the maximization of the PEGR. In terms of this measure, it is clear that in order to achieve pro poor growth we must seek to improve the distribution of national income between poor and non-poor simultaneously with increasing its growth rate.

The issue of the magnitude and pace of poverty reduction can be clarified by distinguishing between two factors that reduce poverty: (1) The first factor is the level of the GDP growth rate. (2) The second factor is enabling a larger proportion of the increased income to go to the poor relative to the non-poor. It is clear that given the first factor, the greater the magnitude of the second factor, the greater is the magnitude of poverty reduction. It is equally clear that if the distribution of income remains unchanged then the magnitude of poverty reduction would depend entirely on the level of GDP growth. Now we come to the tricky part. If the distribution of income becomes more unequal over time as it has done in Pakistan’s case, then as I have argued in these columns, a much higher target GDP growth rate would be required to reduce poverty. This is necessary for an accelerated GDP growth to compensate for the deterioration in income distribution. Under these circumstances if the magnitude of the increase in the growth rate of GDP is smaller than the targeted rate, then in fact such a small increase in the GDP growth rate would be associated with growing poverty. This is indeed what has happened in Pakistan’s case over the last three years: GDP growth has increased and so has poverty.

While Kakwani and Son have made a seminal contribution by providing a precise measure of pro poor growth in their methodology the structure or content of growth remains a black box. Their measurement method would simply indicate whether or not, and the extent to which the GDP growth in a particular period is pro poor. They have not answered the question: What are the specific factors that will change the structure of GDP growth to lead to a higher and more equitable growth? This of course is understandable, because the answer depends on the specific economic features of the country we are dealing with. In Pakistan’s case I have in these columns and in my recent published work attempted to articulate the outline of such a growth strategy. (See for example, Restructuring Growth for Faster Poverty Reduction, DT 3rd April 2003). It would accelerate overall GDP growth and simultaneously enhance its poverty reduction capability.