To continue with the (unfortunate) theme of my last several articles - the disappearance of honest debate in India. The disappearance is bad enough but to replace it with fake news, and/or flaky analysis, is deplorable. Just last week ('Looking for Honesty in all the wrong places', IE, October 14), I had documented how researchers, and even a redoubtable institution like the RBI and the MPC, were indulging in, at a minimum, very questionable and non-reproducible analysis of the effect of fiscal deficits on inflation. (See 'Attempting to Reproduce RBI Results on Fiscal Deficits Causing Inflation' at http://ssbhalla.org/blog). An ill-informed remedy can only aggravate a patient's illness, not make her get well.

Joining the club of the slightly less-than-honest analysis and prescription is IFPRI, with its changing hunger index for the world, and India. It used to be the case that "in the name of the poor" institutions and scholars talked about how poverty in India was high, and climbing. That appeal became weak when absolute poverty in India in 2011/12 came out to be around 12 per cent, not 23 per cent as officially reported (Tendulkar poverty line). The difference between 23 and 12 per cent is the difference in measurement of consumption measured on a 30-day recall basis for food rather than the more accurate seven-day recall basis. The latter is to be preferred, as most economists, including Nobel laureate Angus Deaton, have advocated.

In this ultra-competitive world, you need a peg. I ask you, how many of us would have heard of the IFPRI - the organisation or its hunger index - if it hadn't been about the story that appeared, not coincidentally, in the newspapers about a little girl dying of "hunger". Helped by the IFPRI's misinformed and faulty analysis, in the name of the poor, povertarians are back with a bang.

On a cursory - let alone close - examination, the IFPRI hunger index is not a hunger index at all; it is an index about child mortality, and stunting, and wasting, and undernourishment of children. Improvement in these indicators is an important social goal but it is important that we identify the proper causes so that a proper remedy can be prescribed. Otherwise, organisations run the risk of being ridiculed, and dismissed, as mere publicity seekers.

A definition of hunger is needed to evaluate policies to alleviate hunger. The conventional approach is to measure hunger via calorie consumption. This approach has a long history and follows from first principles. Hunger, by definition, is lack of food. The most basic form of food is calories. Therefore, the reasoning is that lack of food is a good proxy for lack of calories. The chain of IFPRI's reasoning is as follows: Hunger = poverty = lack of food = low consumption of calories. So, the definition of hunger, and counting and identification of the poor, narrowed down to the counting of calories.

However, the caloric equation (poverty means low consumption of calories) has long been given up as an indicator of hunger, or much else. Over 80 per cent of Americans were found to be malnourished and consuming too few calories in 1973 (much below the FAO 2,400 calories per day line), and hence poor. Casual empiricism, and detailed research, has shown that too many calories - obesity - is a major problem in the US, not under-nutrition. Indian scholar P.V. Sukhatme documented as early as 1973 that most of the variation in the consumption of calories was due to genetics.

As part of the regular consumer and expenditure surveys, the NSSO has been regularly collecting data on the magnitude of hunger in India. The exact wording of the NSSO hunger question in (2004/05) was: "Do all members of your household 'get enough food every day: Yes, every month of the year-1; some months of the year -2; No month of the year-3". In 1983, some hunger was reported by 14.2 per cent of the population but by 2004/5, this fraction had declined to only 1.4 per cent, that is, "no" hunger, at least according to consumption surveys on self-identified hunger. Year 2004/5 is the last year the question on hunger was asked in the NSSO survey. A decade later, the Indian government passed a Right to Food law to alleviate hunger. There is an alternative survey measure of hunger for India (and most countries of the world) provided by Gallup surveys. The question they have asked since 2007 is the following: "Does your family have insufficient money/finances to meet food needs?"

The table documents various statistics about hunger and poverty in the world around 2013. Two separate indices are reported for the IFPRI index - one for 2014 and one for 2015. Note that for most regions (except South Asia and India) the new IFPRI hunger index (because of separate indices for stunting and wasting rather than one joint index of underweight) has near double the proportion of hunger.

According to the Gallup hunger index, about 22 per cent of the South Asian population suffers from hunger, somewhat more than the corresponding 32 per cent for the three-times richer East Asian nations. As much as 11 per cent of the population in advanced economies reports itself as hungry. As discussed in my paper for Brookings (see note to table), it does appear that hunger is more of a relative, than absolute, concept.

Malnutrition affects stunting and weight and despite having considerably higher per capita income, India has the same IFPRI nutrition (reported as hunger) status as sub-Saharan Africa. There is a genuine nutrition absorption problem in India. The most likely cause of this is bad sanitation, a large component of which is open-defecation. Arvind Virmani ("The Sudoku of Growth, Poverty and Malnutrition: Policy Implications for Lagging States", Planning Commission 2007) was the first to highlight the importance of sanitation in determining nutrition status. "For instance a child suffering from diarrhoea much of the time is unlikely to be able to ingest much good and healthy food and absorb the nutrition, even if it is freely available and provided to the child by the mother/parents. In the Indian environment, access to water and toilets, breast feeding (to impart immunity in an unhealthy environment), access to sound health advice/treatment, prevalence of vaccination and availability of vitamin supplements" are indicators of bad health, malnutrition etc. (emphasis added).

Some years later, Dean Spears (2013), "The nutritional value of toilets: How much international variation in child height can sanitation explain?", documented the important role that open defecation played in explaining India's bad indicators relative to sub-Saharan Africa. In 2014, Prime Minister Narendra Modi started the drive to stop open defecation, and said that we needed to do so from the ramparts of the Red Fort. We can debate endlessly about the speed of adoption of toilets, but there is now a Bollywood movie on the subject. History will record, a few years from now, the large role played by toilet construction, and adoption, in increasing nutrition efficacy in India to East Asian levels.

I want to end with the following question: Would India have not been considerably better off in terms of health, nutrition and welfare if instead of spending thousands of crore on food subsidies, it had spent monies on essential public goods like sanitation?