The lack of quality data has long plagued Indian policymaking. The error is most acute after a lockdown overnight last year left countless immigrants stranded in our cities without adequate livelihoods. Millions brave fatigue, hunger and accidents — some run by train — as they drive highways and railway tracks to their villages hundreds of kilometers away. It was a huge tragedy that shook our conscience. The government’s delayed response to not being able to identify all those whose aid was delayed has been somewhat pinned down. We do not have a database on migrant workers. Indian Labor Minister Santosh Gangwar agreed in Parliament last September that no information was available on how many migrant workers had died and how many had left urban India. Numbers collected from the states later gave us an estimate of 10 million returnees, although some say that number is much higher. A similar statistical scattering attends to official snapshots of employment. In an effort to address India’s glaring data gaps, our Ministry of Labor last week announced two large-scale surveys. Properly covered, it will withstand a great deal of adverse conditions.
The first of those studies will survey approximately 300,000 households to get a picture of our immigrant population, from our urban population to their living conditions and livelihood resources. Another, a quarter, aims to collect job creation data from 150,000 companies. These are large sample sizes, and while we need reliable numbers on daily-wage editors and job scenarios in the country to assist policy-formulation and welfare interventions, the most important is not scale, but accuracy. We need to get this right right from the start. Details of this initiative are not yet available, but survey designers should ask themselves a few questions before hitting this field. One, if a parallel exercise is performed independently, will it give the same results? Two, is the model really random, faceless and representative? As far as possible, will everyone in the target group be included in it equally? And are all the questionnaires neutral to keep the biases brought about by the questions? Beyond that, definitions also need clarity, because most of our statistical cracks confuse what qualifies. Unemployment in disguise, for example, makes it difficult to define a ‘job’.
Living standards have long been tracked by the National Sample Survey Office (NSSO), but one of its reports of declining household consumption in 2019 led to controversy. That episode spawned a toss-up: either the survey was flawed, or the center’s relationship with reality. A few years ago, when the Center revised how our national income should work, allegations arose that the Indian economy was being clothed. This has weakened confidence in our official data. However, none of our statistical alignment is on damn-spot levels, certainly, beyond correction. Also, some snapshots in India are too slow for policy use. For example, NSSO’s employment survey is rarely done, with analysts relying on the tracker running the Center for Monitoring Indian Economy. The launch of two new mega surveys should prompt our government to review its statistics, lay out all its methods for public scrutiny and address all that is necessary for fixing.