If you walk into any village and ask people who are its poor residents, it is not difficult for them to answer. They will point possibly to a blind widow, an old couple who get by begging for alms or children and men in debt bondage.
They may list low-caste landless agricultural workers who migrate several months every year to brick kilns in the city, the farmer with a tiny holding of a couple of acres precariously dependent on erratic monsoons and who cannot feed his family for the full year even in a good year, households in which children roll bidis instead of attending school, and the out-of-work weaver.
In a city, people will look for the poor in slums or pavements or construction sites, for rag-pickers, casual workers, rickshaw pullers, beggars or domestic workers.
But when government officials are asked the same question, they invariably flounder and fail. There have been three official national surveys to identify poor rural households — in 1992, 1997 and 2002 — and government’s own evaluations admit that more than half the poor are left out of those they select.
If you are poor, there is a greater chance that you will not find yourself on the government’s list. This chronic inability to identify the poor is because when the government asks who is poor, the question is not disinterested. On its answer hinges whether a household will get subsidised food, or free medicines, or a free house, or a bank loan.
There is, therefore, understandably a clamour everywhere to be included in this ‘magic’ list of the ‘government’s poor’ that opens so many doors. In a society ridden with highly unequal power and a notoriously corrupt and unaccountable local administration, it is not surprising that the lists of the poor that emerge are so deeply flawed.
The problem begins because government estimates of the total number of people who are poor are typically based on faulty assessments of what households consume and spend. But government loses its way completely when it goes beyond the question of how many people are poor, to ask instead precisely which people are poor.
It needs to know this because the government seeks to restrict the numbers of persons who access any public services — such as subsidised food, health care, free housing and cheap credit — to fit into its estimates of poverty.
The first two national surveys in 1992 and 1997 were based on assessments of income and consumption respectively of each household. But it is difficult to estimate incomes in casual and self-employment, coloured further by the bias to under-estimate income and consumption so as to be included as poor.
These criteria were wisely abandoned in 2002, for socio-economic indicators of poverty. But many of the actual 13 indicators adopted by the Planning Commission in its third national rural survey of poor families were subjective and somewhat whimsical. A household was in peril of being regarded as relatively well-endowed and consequently ineligible for subsidised food and other government aid if the home had a pucca roof, a toilet, children attended school and some of its members were educated, accessed credit in times of need and occasionally ate non-vegetarian food.
The survey disqualified hunting and foraging forest-based tribal and fisher folk, or conscientious poor people who benefited from government sanitation drives or sacrificed a great deal to send their children to school.
The central government is at the verge of launching its fourth national survey of rural poor households. Since it has also promised to pass a law which guarantees each of these households 35 kilograms of rice at R3 every month, wheat at R2 and millets at R1, it is even more urgent that it gets the survey right this time.
On its outcomes rests the chances of survival with dignity of the poorest families otherwise condemned to live with hunger and uncertainty, by enabling them to access official food and safety nets. Because of government’s failures to identify the poor, the best answer is to offer food and pensions to all households who seek it.
But if universalisation is not acceptable, government must find far more reliable ways to identify the poor.
We propose that we begin by excluding from the survey the richest rural households: those with large irrigated farms; or with three or four wheeled motorised vehicles; mechanised farm equipment, such as tractors, power tillers, threshers and harvesters; or a salary of over R10,000.
Next, we propose the automatic inclusion of all households who belong to certain clearly defined socio-economic categories known to have high levels of poverty, the kinds of households villagers themselves identify when asked who are poor. It is difficult to determine which household has low income or consumes too little.
But there will be far less ambiguity if we ask who the widows are, or disabled, or destitute or bonded workers. These are groups in which the majority of households are poor and dispossessed. Even at the risk of including some who are less deserving, one can ensure that the neediest are not left out if all such social categories are included.
This is a ‘social inclusion’ strategy.
We believe that the next highest priority should be given to Scheduled Castes and Scheduled Tribes which lag far behind most other populations in terms of most socio-economic indicators, followed by Muslims, who share development deficits almost on par. Landless workers, homeless persons, those with a single katcha room, and those with a literate adult above the age of 25 would also be prioritised.
Veteran development expert Robert Chambers often remarks that public officials make wrong decisions because we do not consult the greatest poverty experts of all — poor people themselves. If we listen to our people, they will tell governments how to find the poor among them.
(Harsh Mander is director, Centre for Equity Studies)
(The views expressed by the author are personal)