The curious case of poverty line

My stint as a professional economist lasted for a year in 1964, spent in Dhaka university as a teacher after which I switched over to government service, a career move that had become commonplace given prevailing social values. In my various incarnations in the new profession I had occasions to be intermittently engaged with economics mostly in its applied form of planning, making inputs to policy making and as an implementer of decisions taken. One of the areas that had bearing with the discipline of economics that recurred in various stages of my career as a civil servant was poverty alleviation. Bangladesh being a 'test case of development' it was natural that poverty in its various dimensions would be treated by governments of all stripes with priority and a sense of urgency. The various donors eager to give aid and loan for poverty alleviation acted as a spur to elevate the momentum of government efforts to higher level, albeit incrementally.
The first task in respect of poverty alleviation was to find out who were the poor. This required definition and measurement of poverty. While studying abroad in 1960 the first definition of poverty that I came across was the tantalising remark by Ragner Nurkse that a person is poor because he is poor. He, of course, elaborated this pithy observation with his exposition of the 'vicious circle of poverty'. But it did not establish the poverty line or postulated the method of measuring poverty. Michel Lipton's Why Poor People remain Poor? (1977) was an explanatory exercise that focused on urban bias for prevalence of rural poverty. The article by Paul Streeten Poverty Concepts and Measurement that appeared in the Bangladesh Institute of Development Studies' (BIDS), special issue on poverty in 1990 was a step forward towards establishing the poverty line but failed short of specifics. 
The earliest work on establishing a poverty line for measurement of poverty was done by A. R. Khan and whose consultancy work with the ILO in 1975 produced the paper titled, Poverty and Landlessness in Rural Asia. He used calorie as the measurement of poverty and defined absolute poor as those who could not take more than 80 per cent of the required calorie intake of 1935 Kc. According to this measurement criteria the extremely poor were those whose calorie intake was 80 per cent of the required 1720 Kc. It is obvious that only food items were considered by Khan to establish the poverty line and estimate the percentage of households and populations as either absolute or extremely poor, depending on food intake. The quantitative nature of the measurement in terms of calorie intake was slightly improved by a study in early Eighties by applying the cost of normative minimum dietary pattern as recommended by the Food and Agricultural Organization (FAO) to the pattern of Quarterly Survey of Current Economic Condition (QSCEC) and the Household Expenditure Survey of the Bureau of Statistics. Two changes appeared in this study conducted by Q. K. Ahmed and M. Hossain in 1984. Firstly, the minimum per capita daily consumption of food was raised by FAO to 2332 calorie (11 per cent higher than the average calorie intake estimated in the 1976-77 Nutrition Survey and the minimum dietary pattern was monetised to arrive at the cost of food consumption. The important thing was that instead of only calorie requirement as a measure of poverty line the cost to cover the dietary pattern was estimated which led to the income criteria underlying poverty line. But it was an incomplete measurement as the FAO estimate was entirely based on food consumption excluding non-food items. 
The World Bank estimated poverty in 1987 using Household Expenditure Survey data of BBS and a consumption bundle based on the actual intake by the poor. Though the consumption bundle included non-food items and the expenditure survey corresponded to this, the determination of Basic Needs Income was not done. This was the Achilles heel of all the estimates and measurements of poverty up to 1987. That the starting point of defining poverty and drawing a poverty line required determination of basic needs to remain above poverty eluded the researchers for reasons best known to them. Had they realised the importance of this the next and the very important steps determining the Basic Needs Income (to meet basic needs) would have drawn their attention. The Basic Needs Income would have been the Holy Grail that they subconsciously sought but failed to articulate. It was perhaps natural in a branch of intellectual enquiry that was new and had no precedents to go by. 
It was perhaps UNDP (United Nations Development Programme)  that pointed to the concept of basic needs and Basic Needs Income (BNI). Under the basic needs approach the bundle of consumption included both food and non-food items, particularly housing, health, education and transport. While BNI estimated the income required to have access to the consumption bundle. The BNI became the poverty line with which poverty could be measured on the basis of the Quarterly Survey of Household Income and Expenditure Survey of BBS. Using the basic needs approach the BIDS initiated the analysis of poverty trends in 1989 to establish a national representative baseline which would present a continuous monitoring of poverty over time. But the study enlarged the scope of basic needs by adding vulnerability, coping capacity and social attitudes to the tangible basic needs goods and services that could be quantified and monetized (Z. Rahman and Hossain 1995). Since no proxy value was computed for the non-tangible needs (vulnerability, coping capacity) the basic needs income used for generating panel data i.e. the tracking of the same set of representative households over time led to the establishment of the national baseline in 1989 covering 1300 rural households in 62 villages which was followed up by continuous re-surveys on a sub-sample of 350 households in 17 villages at annual intervals. The latest resurvey that is within my knowledge goes back to 1995. It is quite possible that this important survey of poverty situation was discontinued by BIDS with the stoppage of assistance by the donor (World Bank?) as is normally the case with most research undertakings and development projects. My conjecture is supported by the fact that a new study on poverty indices based on available household estimates was undertaken in 1992 by BIDS where the basic needs approach was used (Hossain and Binayek Sen, 1992). In 1995 an independent study was undertaken by Sen at the initiative of  the Centre For Policy Dialogue using the same methodology, more or less. 
Similar poverty studies with similar methodology might have been undertaken in other countries establishing country-specific poverty line and measurement of poverty on the basis of the same. Given this rich body of research and establishment of national poverty lines in different countries what was the compulsion for the World Bank to come up with a global poverty line? Granted, poverty is a global phenomenon, but it is country-specific and as Tolstoy said famously, 'All happy families are alike, all unhappy families are unhappy in their own way', similarly poverty cannot be homogenised applicable to all countries using purchasing power parities (PPP). Moreover, there is no overwhelming need for this to address the problem of poverty which is done best for each country on the basis of its own poverty line and measurement. In spite this being redundant and unnecessary, painstaking efforts are being made in a burgeoning department that has sprung up in the World Bank where hairs are being split whether global poverty line can be estimated by US $1.88 or $1.90. Not being content with its homegrown work, the Bank has now established a 24-member Commission on Global Poverty which will submit its report next spring. Instead of these types of academic exercise which end in futility, poverty analysis in the form of updating poverty line and measurement of urban and rural poverty should be left to individual countries which have by now adequate expertise in this area. 
Before concluding, I feel tempted to quote from an article by a former chief economist of the World Bank, 'We ran two experiments; one was to inflate the poverty lines of 15 countries used in 2005, using their respective inflation rates and then taking an average; the other was to do the same for 101 countries for which we had the necessary data. Since there had been inflation between the two rounds of the PPP computation, in 2005 and 2011, we would obviously have to raise the nominal poverty line to keep the real line constant. However, doing this for the world as a whole is far from easy.' (Kaushik Basu, Project Syndicate, FE January 26, 2017). One cannot but sympathise with the backbreaking work that the World Bank staff have undertaken for the establishment of a global poverty line. But pray, why try this? Where is the compelling need? The poor countries are poor no more in matters of poverty analysis, for which the Bank can take some credit. Why not let them be self-reliant in this respect? If I sound like an ignoramus I beg to be excused. After all, I left the profession of economics long ago and am not familiar with all the arcane rituals that the discipline has come to assume to mystify ordinary mortals for self-preservation. But my common sense has not left me and on that I rest my case.                          
[email protected]